[lcd4ever]

Still cannot decide which Blu-ray player to upgrade to...I currently use a Sony BDP -S550.....

My main reason for upgrading is, DVD upscaling ,feel I can squeeze even more out of my DVD collection....

For display I am using the Sony KDL-52X2000 panel

Right from what info I have gathered on these forums the 2 progressive scan engines that give the best upscaling performances are the Anchor Bay VRS, as used by the following shortlisted players.....
Oppo BD-83
Marantz BD 7004
Denon 2010......these players are all around the £600 mark...
Now the Realta engine seems to be favoured by the most expensive players eg the new Denon AU1, the Marantz 9004 these players are coming in at £4000 way over my budget......but I could stretch to the
Denon 3800 BD(Realta engine) at its new price of £800 if it is worth it....

OK so to summarise I know Blu-ray pictures = very little difference as no processing(unless you want it) by the player, Upscale DVD virtually all down to the player this is the area I am looking to see the biggest improvement over my Sony in.......so which of the above players will do it...?

Can anyone 'objectively' tell me what the difference is in the processing by these scan engines, ie what visible differences and why ? Surely the three AB models short-listed will deliver the 'same' upscaled picture if not ....why ?

To clarify before the 'Oppo brigade' chime in speed of loading and SACD/DVD-A are not factors here I want THE best picture from my discs for <£2000.....!!!

All audio will be handled by my Yamaha RX-V1900 receiver via HDMI so analogue out performance not important either...

Hopefully a post here will finally make my mind up....

[David Mackenzie]

Right from what info I have gathered on these forums the 2 progressive scan engines that give the best upscaling performances are the Anchor Bay VRS, as used by the following shortlisted players.....
Oppo BD-83
Marantz BD 7004
Denon 2010......these players are all around the £600 mark...
Now the Realta engine seems to be favoured by the most expensive players eg the new Denon AU1, the Marantz 9004 these players are coming in at £4000 way over my budget......but I could stretch to the
Denon 3800 BD(Realta engine) at its new price of £800 if it is worth it....

Very few reviews actually talk about the Scaling quality, because this can be somewhat subjective and hard to quantify. When people talk about upscaling performance, they're often not talking about scaling at all, but deinterlacing.

But to be brutally honest: most of them are going to look the same with the majority of DVD titles, which don't really have a lot of high frequency information to start with. It would take a seriously bad scaling algorithm to add any more ringing and/or detail loss to most DVDs.

[Avi]

Originally Posted by lcd4ever

OK so to summarise I know Blu-ray pictures = very little difference as no processing(unless you want it) by the player, Upscale DVD virtually all down to the player this is the area I am looking to see the biggest improvement over my Sony in.......so which of the above players will do it...?

Can anyone 'objectively' tell me what the difference is in the processing by these scan engines, ie what visible differences and why ? Surely the three AB models short-listed will deliver the 'same' upscaled picture if not ....why ?

To clarify before the 'Oppo brigade' chime in speed of loading and SACD/DVD-A are not factors here I want THE best picture from my discs for <£2000.....!!!

As David points out people often use the word "scaling" to mean the whole video processing chain not just resolution interpolation. I also agree with him that some aspect of video processing in particular the resolution interpolation processing and image enhancement can be more about subjective preference rather than technically better. I use the analogy that it's can be a bit like adding MSG to food to enhance the taste.

Personally I look at five main areas of performance with DVD which are 1) Mpeg decoding, 2)PAL & NTSC cadence deinterlacing, 3) resolution interpolation, 4) reference output levels i.e. chroma, luma, no clipping etc. and 5) output options i.e. colour space. There is a wide range of test material that enables you to see how a player performs in these area under controlled conditions. Whilst some of these condition are unlikely to exist on every title you watch if they do the test shows how the player is likely to perform. Some conditions such as chroma resolution may affect all output and similarly scaling algorithms that exaggerate edge acutance may be more noticeable as the image size increases.

HERE is a link which compares aspects of Realta to the ABT processor in Denon/Oppo products using third party test material. Silicon Optix did produce a suite of tests material called HQV that was aimed at showcasing the performance benefits of their HQV processors. The HQV processors usually passed the HQV tests yet could fail similar independent tests so it perhaps important to use a range of material.

AVI

Edit

I've added a couple of photos showing how different scaling algorithm produce different results. The use of increased edge acutance is indented to create the illusion of increased sharpness/detail. One of these images is from a current top of the range Blu-ray player and both are using 0 sharpness/enhancement settings. These are easier to observe with test material and may not be perceived as obviously using moving video but it still exists.

[Ian_S]

Originally Posted by lcd4ever

Still cannot decide which Blu-ray player to upgrade to...I currently use a Sony BDP -S550.....

My main reason for upgrading is, DVD upscaling ,feel I can squeeze even more out of my DVD collection....

For display I am using the Sony KDL-52X2000 panel

Right from what info I have gathered on these forums the 2 progressive scan engines that give the best upscaling performances are the Anchor Bay VRS, as used by the following shortlisted players.....
Oppo BD-83
Marantz BD 7004
Denon 2010......these players are all around the £600 mark...
Now the Realta engine seems to be favoured by the most expensive players eg the new Denon AU1, the Marantz 9004 these players are coming in at £4000 way over my budget......but I could stretch to the
Denon 3800 BD(Realta engine) at its new price of £800 if it is worth it....

OK so to summarise I know Blu-ray pictures = very little difference as no processing(unless you want it) by the player, Upscale DVD virtually all down to the player this is the area I am looking to see the biggest improvement over my Sony in.......so which of the above players will do it...?

Can anyone 'objectively' tell me what the difference is in the processing by these scan engines, ie what visible differences and why ? Surely the three AB models short-listed will deliver the 'same' upscaled picture if not ....why ?

To clarify before the 'Oppo brigade' chime in speed of loading and SACD/DVD-A are not factors here I want THE best picture from my discs for <£2000.....!!!

All audio will be handled by my Yamaha RX-V1900 receiver via HDMI so analogue out performance not important either...

Hopefully a post here will finally make my mind up....

Your biggest issue with Blu-ray that I can see is infact your screen.

The X2000 series doesn't support 24fps, so you're already taking a downgrade as ideally you need to feed your screen 1080p/60 from Blu-ray, which introduces more judder.

The new Denon 2010, whilst it has VRS, has a lesser version that only works for SD material, not HD.

Hard as it will be to take, rather than spending up to £2k on a BD player, screen technology moves fast and you might get much better PQ improvements by spending £2k on a new screen. 24p support would improve general fluidity (although it won't remove judder from camera work inherent to the source) and if you have many R1 NTSC DVD's, would allow 1080p/24 DVD output too with a suitable BD player like the Panasonic BD60 if you use HDMI audio.

[dj_london]

Hi,

Hopefully I'm not hijacking the thread but I have a related question. How does the the image quality for DVD playback compare using a Blu Ray player as opposed to a dedicated DVD player. My reasoning being that it could be better to buy a good second hand DVD player for DVD's and leave the Blu Ray player.

Cheers,
DJ

[Avi]

A couple more photo examples that attempt to show the subtle difference deinterlacing can cause to the end image with real world material. In reality these artefacts may only appear briefly depending on the title and scene content. However, it's often this level of performance that differentiates solutions rather than night and day issues. In this case both use the same chip but the available deinterlacing options are different due to how the chip features have been implented in the two different products.

AVI

[David Mackenzie]

I agree with Ian_S - I'd upgrade the display first if you really want a picture quality improvement.

[lcd4ever]

Display upgrade is definitely a resounding no-way; I paid nearly £4000 for this TV only 2 years ago, and I still have to see it bettered, I have a newer W-series 4000 in the bedroom which is not a patch on the X-series.....so I will be keeping this TV till it dies of natural causes......
I'm very happy with the Blu-ray picture from the Sony and I know if I get eg the Denon 3600 I will get only a slight improvement in Blu-ray picture; but I know from first hand experience that upscaled DVD improvements can be significant; I have a Panasonic DVD recorder with upscaling which I thought was great to begin with then I bought the PS3 and it was another significant improvement, the the BDS-550 was another improvement albeit smaller so looking for just that little bit more yet...

Thought about the separate DVD player route looked into the Arcam DV139(2nd hand can be got for around £700) but it uses the Anchor Bay chip-set.....
If Denon/Marantz put the Realta chip in their £4 K players just wondering if it is really that good now it can be had in the 3600 for £800 which I don't mind spending.....

[Avi]

Originally Posted by lcd4ever

If Denon/Marantz put the Realta chip in their £4 K players just wondering if it is really that good now it can be had in the 3600 for £800 which I don't mind spending.....

I think you may be surprised by the cost of the video processing chip as it is relatively cheap.

Here's the original 2004 announcement from Silicon Optix (who no longer exist and the Teranex Realta IPR is now owned by Jupiter Systems and the Reon by IDT)

"Silicon Optix Unveils ´Realta´ Chip
By Greg Tarr -- TWICE, 9/20/2004

INDIANAPOLIS — Silicon Optix, a San Jose, Calif.-based silicon chip manufacturer, used CEDIA Expo to announce a video processing IC it has developed with Teranex, one of the leading suppliers of imaging processing solutions for the broadcast and post-production industries.

Silicon Optix called its IC the “Realta chip.” It combines Teranex´s trillion-operation-per-second broadcast-quality video processing with Silicon Optix´s proprietary geometric-scaling technology to create a very high performance standard for up-converted HD and SD image sources, which the company is calling “Hollywood Quality Video” or HQV.

Realta´s programmability allows consumer electronics manufacturers to optimize image quality for different types of display technologies such as plasma, LCD, DLP, HTPS, D-ILA and LCoS.

Silicon Optix, which recently acquired Teranex, is recruiting video display manufacturers to adopt its new video processing circuitry for future products.

Although Teranex´s systems are known to cost as much as $65,000 in professional applications, Silicon Optix executives said they were able to tap many of the same benefits for the Realta chip, which will carry a minimal added cost when incorporated into one of today´s higher-end displays.

“Leading up to the acquisition, Silicon Optix has been working with Teranex under a veil of secrecy for almost two years,” said Dennis Cresp, Silcon Optix´s marketing VP."

AVI

[lcd4ever]

Thanks AVI I realise that the cost of most of the incorporated chips in these products is relatively cheap, I was just wondering why Denon put the AB VRS in the new 2010 but the Realta in the AU1 and the 3800; they must feel it gives the superior image......yet going on the test results you gave me the link to the AB seems to out-perform it in every way......I may e-mail Denon and ask them this......

In your own personal opinion which chip gives the superior image ? Really asking you if you think the 3800 is a good buy at £800 or would I be better to save 200 quid and go for one of the other players......which would you choose from those short-listed and why ?

Thanks again

[Avi]

Originally Posted by lcd4ever

In your own personal opinion which chip gives the superior image ? Really asking you if you think the 3800 is a good buy at £800 or would I be better to save 200 quid and go for one of the other players......which would you choose from those short-listed and why ?

The chip is just one albeit important part of the equation. There are other factors such as the display and calibration that can be as equally important in terms of achieving reference image quality.

It also really depends on what you mean by "superior image" as my reference may be different to yours plus our setups are likely to be different. Thus what I would choose and why may be very different to you and probably wouldn't help.

Personally I think the 3800 really comes into its own when MC analogue output is a key requirement due to bass management etc. If that's high on your list then it's a very strong contender. If it's not then I wouldn't get too hung up on Realta v ABT etc.

AVI

[StooMonster]

Also the default setting they put on each chip, the OEM's don't just plug and go.

ABT has it's strong points, it's probably the best at locking on to weird cadence such as Japanese Anime but can get tripped up on difficult 2:2 locks.

Also to consider, OEM's don't always want "the best" they want the best price/performance and will go shopping with a list of boxes to be ticked in regard to features.

StooMonster

[cpcat]

Originally Posted by Avi

HERE is a link which compares aspects of Realta to the ABT processor in Denon/Oppo products using third party test material. Silicon Optix did produce a suite of tests material called HQV that was aimed at showcasing the performance benefits of their HQV processors. The HQV processors usually passed the HQV tests yet could fail similar independent tests so it perhaps important to use a range of material.

.

I would temper those comments a bit by saying that ANY test disc is prone to the same. The S&M disc referenced in the link is just as prone to the same sort of bias and indeed is provided "free" with the new Oppo BD player using the ABT2010 chip.

My testing b/w a ABT2010 based VP (DVDO Edge) and a Realta based VP (Denon DVP-602ci) yielded differing results. See Denon Video Processor - Page 2 - AVS Forum.

Two advantages to Realta are without room for debate. One is the presence of temporal noise reduction (absent in ABT2010) and two is superior scaling (more ringing w/ABT2010).

My testing reveals superior video deinterlacing for 1080i w/Realta (about equal to ABT2010 with 480i). For film, testing (both 480i/1080i) was about equal. However, with real-world performance Realta is clearly superior in producing usable, artifact-free 1080p24 from film sources (480i/720p/1080i both pre-recorded and broadcast) while ABT2010 stumbles especially with broadcast material producing unacceptable jerkiness. This tells me that Realta picks up the cadence more rapidly and accurately more of the time vs. ABT2010.

[lcd4ever]

Good to hear as I went with the Denon 3800.......

[mikelj]

Originally Posted by cpcat

I would temper those comments a bit by saying that ANY test disc is prone to the same. The S&M disc referenced in the link is just as prone to the same sort of bias and indeed is provided "free" with the new Oppo BD player using the ABT2010 chip.

My testing b/w a ABT2010 based VP (DVDO Edge) and a Realta based VP (Denon DVP-602ci) yielded differing results. See Denon Video Processor - Page 2 - AVS Forum.

Two advantages to Realta are without room for debate. One is the presence of temporal noise reduction (absent in ABT2010) and two is superior scaling (more ringing w/ABT2010).

My testing reveals superior video deinterlacing for 1080i w/Realta (about equal to ABT2010 with 480i). For film, testing (both 480i/1080i) was about equal. However, with real-world performance Realta is clearly superior in producing usable, artifact-free 1080p24 from film sources (480i/720p/1080i both pre-recorded and broadcast) while ABT2010 stumbles especially with broadcast material producing unacceptable jerkiness. This tells me that Realta picks up the cadence more rapidly and accurately more of the time vs. ABT2010.

Although it probably is worth pointing out that you were using the EDGE.

Products using the same ABT210 chip may have different performance depending on how the processing capabilites of the ABT2010 chip have been implemented. For example, the EDGE does not have the 2-2 odd/even modes that other products using the ABT2010 have, e.g. Oppo 83/983, VP50 Pro, so 2-2 de-interlacing in the EDGE can be bettered by other products using the ABT2010.

[Avi]

Originally Posted by cpcat

I would temper those comments a bit by saying that ANY test disc is prone to the same. The S&M disc referenced in the link is just as prone to the same sort of bias and indeed is provided "free" with the new Oppo BD player using the ABT2010 chip.

My testing b/w a ABT2010 based VP (DVDO Edge) and a Realta based VP (Denon DVP-602ci) yielded differing results. See Denon Video Processor - Page 2 - AVS Forum.

Two advantages to Realta are without room for debate. One is the presence of temporal noise reduction (absent in ABT2010) and two is superior scaling (more ringing w/ABT2010).

My testing reveals superior video deinterlacing for 1080i w/Realta (about equal to ABT2010 with 480i). For film, testing (both 480i/1080i) was about equal. However, with real-world performance Realta is clearly superior in producing usable, artifact-free 1080p24 from film sources (480i/720p/1080i both pre-recorded and broadcast) while ABT2010 stumbles especially with broadcast material producing unacceptable jerkiness. This tells me that Realta picks up the cadence more rapidly and accurately more of the time vs. ABT2010.

That's why I said it's important to test with a range of material.

I think you will find I'm also not a fan of ABT's scaling and have commented on this but that said I'm also not a fan of Realta scaling either. I also agree that the noise reduction and image enhancement is more crude on the ABT. That said the Realta doesn't support PReP which means it can't help with many products i.e. SkyHD that are poor at deinterlacing and don't support interlaced output over HDMI. I find noise reduction most useful with poor Sky/broadcast material and don't use it with DVD. So it really is a case of choose your poison based on your requirements as PReP wouldn't be beneficial in a DVD player but it would in an AVR/VP etc. I'm not hung up over any solution as they all have relative good/weak points and non are perfect.

How much testing do you do with PAL DVD ? How does the Realta perform with Finzel or Merighi cadence and motion tests ? I don't think these are linked to a manufacturer but as I said I would use a range of material not just what the manufacturer developed to show case their product i.e. HQV.

AVI

[roman-r]

Can I jump in and ask if PREP definately does what it says on the box i.e. can it really work out 100% of the time which order were the original fields?

nagging at me a little as I've had to start using it as I swapped DVD Player which only has Progressive out of HDMI.

[Avi]

Originally Posted by roman-r

Can I jump in and ask if PREP definately does what it says on the box i.e. can it really work out 100% of the time which order were the original fields?

nagging at me a little as I've had to start using it as I swapped DVD Player which only has Progressive out of HDMI.

It certainly makes a significant improvement with SD material from a Thomson SkyHD box. I'm not sure any video processor claims to get everything right 100% of the time but PReP appears pretty accurate based on my observations. Other video processing solutions I've tried in this capacity i.e. Realta, VXP etc have been unable to resolve issues with material that has already been deinterlaced. The other alternative is a new DVD player that outputs SD interlaced signals over HDMI.

AVI

[roman-r]

Originally Posted by Avi

The other alternative is a new DVD player that outputs SD interlaced signals over HDMI.

AVI

You should never have said that

Guess I need to check the one fine combing atrifact , from film source, that I've noticed with my new setup and see if it is there with old setup. Old setup was component interlaced into VP50 from a Denon A11. Hopefully it will be, so at least I'll give myself breathing space before I go on the search for the "Holy Grail"

A11 with VP50 passed the Gladiator test but that was with forcing correct 2:2 even/odd as otherwise it failed on Auto.

Sorry to Hijack thread.........getting carried away

[cpcat]

Originally Posted by Avi

How much testing do you do with PAL DVD ? How does the Realta perform with Finzel or Merighi cadence and motion tests ? I don't think these are linked to a manufacturer but as I said I would use a range of material not just what the manufacturer developed to show case their product i.e. HQV.

AVI

None, sorry.

It's worth noting (with regards to HQV test material) that for some time it was either the only or at least among the only widely available and decent source for objective display testing. Sure, it may have been developed by Silicon Optix to showcase their chips, but there also just weren't many other options available.

I find it a little interesting that "HQV test material showcases Reon/Realta" gets mentioned so much but no one seems to point out that the same thing goes with Oppo/ABT and the S&M disc. Afterall, they obviously all three have a business relationship.

[cpcat]

Originally Posted by mikelj

Although it probably is worth pointing out that you were using the EDGE.

Products using the same ABT210 chip may have different performance depending on how the processing capabilites of the ABT2010 chip have been implemented. For example, the EDGE does not have the 2-2 odd/even modes that other products using the ABT2010 have, e.g. Oppo 83/983, VP50 Pro, so 2-2 de-interlacing in the EDGE can be bettered by other products using the ABT2010.

I wasn't aware of the differing capability b/w Edge and the Oppo player. Thanks for the info. Obviously that would make a difference in what I was seeing. Edge failed all of the variable cadences in the HQV disc while Realta did fine.

[Avi]

Originally Posted by cpcat

None, sorry.

It's worth noting (with regards to HQV test material) that for some time it was either the only or at least among the only widely available and decent source for objective display testing. Sure, it may have been developed by Silicon Optix to showcase their chips, but there also just weren't many other options available.

I find it a little interesting that "HQV test material showcases Reon/Realta" gets mentioned so much but no one seems to point out that the same thing goes with Oppo/ABT and the S&M disc. Afterall, they obviously all three have a business relationship.

I'm also a little cynical with regards to most thing AV and I think that's healthy. I'm not saying the HQV test isn't useful only that I'd be surprised if a HQV solution didn't pass all the tests. That's the reason I believe it's important to use a range of tests/real world material to get some idea of overall performance.

The Spears & Munsil disc was the first I've used that objectively showed a difference in chroma resolution (not just with the Oppo) using 1080p/24 Blu-ray between different player/display combinations and output settings.

AVI

[PE06MCG]

Originally Posted by mikelj

Although it probably is worth pointing out that you were using the EDGE.

Products using the same ABT210 chip may have different performance depending on how the processing capabilites of the ABT2010 chip have been implemented. For example, the EDGE does not have the 2-2 odd/even modes that other products using the ABT2010 have, e.g. Oppo 83/983, VP50 Pro, so 2-2 de-interlacing in the EDGE can be bettered by other products using the ABT2010.

The enabling of 2-2 odd/even in the ABT2010 seems to have happened with the Duo also (according to the manual that is).

[Avi]

Originally Posted by SilverFox2

The enabling of 2-2 odd/even in the ABT2010 seems to have happened with the Duo also (according to the manual that is).

It wouldn't surprise me if this feature was also added to the Edge.

AVI

[jgrg]

Originally Posted by SilverFox2

The enabling of 2-2 odd/even in the ABT2010 seems to have happened with the Duo also (according to the manual that is).

My Edge combs quite frequently on reports on the BBC News, but never in the studio. I guess they are from different kinds of cameras, which interlace differently: one is odd first, the other even first?

It doesn't comb if I put the deinterlacer into Video mode instead of Auto. Would enabling the 2-2 odd/even feature stop it combing in Auto?

[Avi]

Originally Posted by jgrg

Would enabling the 2-2 odd/even feature stop it combing in Auto?

No.

These are forced cadence modes that replace auto and can only really be used in situations where the source material uses entirely that cadence i.e. a film DVD. It doesn't work properly with mixed cadence material. In its current format the 2:2 odd/even is under the "film mode" deinterlacing option.

AVI