[Buyer’s Guide] Kaleidescape Quality Compared to Blu-ray & other Digital Sources

[Nathan28]

Here's an interesting thing. You have to suffer the indigeneity of owning a Sony something, something, 7000 projector, to get this. It's the Matrix Reloaded. If you watch the disc, then there is a lot of posterization on the freeway section, to the extent that it is almost a moiré pattern. On K it looks very good. If you've seen the disc, then you can just about see moiré/posterization patterns, but only just. This for me is a major enhancement over disc. For the rest of you, you're probably saying what?

Not all encodes are created equal

 

[Surge]

This is the current marketing material on the front page of K's website. It might seem like we're holding them to a high standard, but I feel that higher standard is essential to the product they market.

The real gains can be made in how the hardware is engineered. The stock power supply is absolute junk and should be upgraded to a linear unit. Even then you won’t see the fullest benefit unless you get a very good one.
I have posted here before that my Apple TV X A/V quality was noticeably better than my K-scape, even with a (Teddy Pardo) linear power supply.

Noise and jitter make a big difference. You’re not getting “pure” 1s and 0s out of a digital player for many reasons which can be easily answered with an internet or AI search.

There isn’t much point in comparing sources or content until you have minimized noise. Clean AC, proper grounding, and then the best power supplies you can afford make a Huge difference.

 

[DieZwei]

Sigh, what happened to the good ol´ "shit in, shit out" rule?...

 

[Rmarci]

The KS does not introduce noise, and I have hundreds of lossless HDMI captures confirming it. In fact, it does the opposite, but I’m hopeful it will be fixed soon.

The Apple TV, haha… This player is far from being bit-accurate and can’t even handle SDR correctly (green push). I think that was fixed a couple of weeks ago on the latest model, though. Dolby Vision is also incorrect on the ATV, with YCbCr clipping and colorspace issues, and CMv4.0 is also broken, with clipped colors above 2500 nits.

This was fixed a while ago, but for years the Apple TV couldn’t even output proper 24.000 fps and instead used 23.976 with dropped frames. Not to mention that a ALL of the Dolby Vision content on iTunes has incorrect colors due to the same static color reshaping they use, although that isn’t a hardware issue.

 

[Surge]

Ah, the fact that you use terms like "bit perfect" tells me you don't understand how digital signals travel between components. Noise is absolutely introduced from one component to another and from power supplies and by the cables picking up RFI/EMI. There is no such thing as "bit perfect" when you are playing back a digital source. You are not stopping to check the accuracy - it's not the same as copying a computer file.
1s and 0s are sent as analog through any cable - there is so such thing as sending a "1" - it's an analog waveform.

I was referring to the modified Apple TV, hence the "X".

If you seriously think the K-scape works well with the stock power supply, you have a lot to learn, my friend.
That's all I will say on this matter. If you are seriously interested in learning, there are plenty of online resources for you.
You guys who claim to be experts in this need to wake up -- the fact that you don't understand that EVERY digital signal carries noise and timing issues, is like saying the Earth is flat. It's so basic it's mind boggling.

Dolby Vision - you should not be using Dolby's tone mapping if you're serious about the best performance out of a display. First of all, you should have a native HDR display so you can see actual HDR. Dolby Vision is fake HDR, and it doesn't compare to Lumagen or (presumably, but I have no experience), MadVR.

 

[thrawn997]

@Surge, you might want to take your own advice and do research on this. You are relying on debunked audio/video myths.

You claim electrical noise subtly degrades digital HDMI picture quality. It does not. Digital receivers don't care about the exact shape of the waveform; they only care if the voltage is above or below a specific threshold at a specific time (representing a 1 or a 0). Digital data either arrives bit-perfect or it fails completely; failure manifests as obvious visual artifacts like dropouts or white sparkles, not a noisy image. There is no middle state. The transport is binary: perfect or broken.

And power supply upgrades are snake oil. For a purely digital device like a Kaleidescape, an expensive aftermarket power supply does not change the 1s and 0s on the wire. The bitstream is identical regardless of whether you're pulling from a stock PSU or a $2,000 boutique unit. Same input; same output.

 

[dinom]

Surge, sorry but you are wrong about this. I’m a EE with 30 years of experience designing high speed SERDES interfaces. You are correct in that the signals traveling over the cable are analog but HDMI uses forward error correction as well as other means to recreate the signal at the receiver. If there are any errors at the display that can’t be corrected 100% the image will break up or disappear completely.

And the suggestion that only linear power supplies are capable of low noise is completely incorrect. A good quality switching supply is more than capable (and in fact is usually required due to high current demands) to power a high performance source device. There are things known as filtering that is applied after the switcher to further reduce the noise.

 

[Rmarci]

His whole reply sounds like all AI BS...

Dolby Vision is fake HDR, and it doesn't compare to Lumagen or (presumably, but I have no experience), MadVR.

This alone proves that you have no clue what you are talking about. MadVR and Lumagen are just algorithmic processes that detect scene cuts and apply tone mapping on the fly, and they will make mistakes. Dolby Vision uses pre-calculated metadata, with manual scene cuts and a trim pass created manually by a colorist on a mastering display.

So what you are basically claiming is that a machine is better than a human eye. You should probably educate yourself before coming here with claims that aren’t supported by any data.

 

[todo]

I have FARAD linear power supplies on my Strato V, Dune Media Player, Uoogos AMB6 and own a Cinemike Apple TV (which people Rave about when they hear it) and off course the Lumagen Radiance Pro which the manufacturer has said has 38 linear power supply stages.
Also have Richard Grey Power Supply (the 350 pound beast) recently updated with the latest power filters - the manufacturer sent me a before and after power output and the measurement graphs show huge improvement in linearity of the power (almost flat now versus jaggy sine wave before).
Linear Power supplies especially FARAD is something which needs to be 'heard' and 'viewed' - that is the ultimate test.
My audio was recently calibrated by a world known audio calibrator - and we AB'd together ATMOS MKV with or without the FARAD on the Dune Player and the difference was astounding - the calibrator who believed that LPMs are snake oil was shocked to 'hear the difference'.
Here is my observation on sound quality of my various players playing the same content:
1. Uoogos AMB 6 - Winner
2. Dune Media Player
3. Stratos V
4. Cinemike Apple TV
The cheap Uoogos AMB6 Plus player is a winner and that is shocking but true.
I was not a believer in clean power but when I received my Richard Grey (which powers every thing in my system) I could 'hear' the sound stage had opened up greatly and that lead me to go down the FARAD route and I am loving it.
I own DLP projector which is run on 240 volts so I dont have any power conditioning on it but I understand that Linear Power supplies help LCOS projectors but not DLP as DLP is digital.
Please 'Hear' a Linear Power supply before discounting it - you may be missing something you will never know.
By the way I am a EE Also.

 

[dinom]

We’re talking about digital video standards not analog audio. But if we are talking digital audio most of what I said stands true.

 

[thrawn997]

I have FARAD linear power supplies on my Strato V, Dune Media Player, Uoogos AMB6 and own a Cinemike Apple TV (which people Rave about when they hear it) and off course the Lumagen Radiance Pro which the manufacturer has said has 38 linear power supply stages.
Also have Richard Grey Power Supply (the 350 pound beast) recently updated with the latest power filters - the manufacturer sent me a before and after power output and the measurement graphs show huge improvement in linearity of the power (almost flat now versus jaggy sine wave before).
Linear Power supplies especially FARAD is something which needs to be 'heard' and 'viewed' - that is the ultimate test.
My audio was recently calibrated by a world known audio calibrator - and we AB'd together ATMOS MKV with or without the FARAD on the Dune Player and the difference was astounding - the calibrator who believed that LPMs are snake oil was shocked to 'hear the difference'.
Here is my observation on sound quality of my various players playing the same content:
1. Uoogos AMB 6 - Winner
2. Dune Media Player
3. Stratos V
4. Cinemike Apple TV
The cheap Uoogos AMB6 Plus player is a winner and that is shocking but true.
I was not a believer in clean power but when I received my Richard Grey (which powers every thing in my system) I could 'hear' the sound stage had opened up greatly and that lead me to go down the FARAD route and I am loving it.
I own DLP projector which is run on 240 volts so I dont have any power conditioning on it but I understand that Linear Power supplies help LCOS projectors but not DLP as DLP is digital.
Please 'Hear' a Linear Power supply before discounting it - you may be missing something you will never know.
By the way I am a EE Also.

When playing an video with Dolby Atmos the media player is acting entirely as a digital pass-through. The player does absolutely zero audio processing, digital-to-analog conversion, or spatial rendering. It simply takes the compressed data packets from the file and shoves them down the HDMI cable.

It is the Audio Video Receiver or processor at the other end of the cable that unpacks the data, renders the Atmos objects, and creates the "soundstage" using its own Digital-to-Analog Converters. If the Ugoos, Dune, and Strato V are all successfully bitstreaming the file without dropouts, the AVR is receiving the mathematically identical file in all three scenarios. An expensive power supply on the transport device cannot rewrite the Dolby TrueHD data to make the soundstage wider.

 

[todo]

I cannot comment on video - audio yes

As a observer and hobbyist - it is my job to put this out even though I might get flamed for it - and I know the technical concepts as my education I am a electrical engineer.

My guess is that to really hear the difference you need to have a high resolving audio system.

It is for others to try this out before discounting it completely - I will bow out of this very important thread from @Rmarci and thanks to him for helping to improve the video on K.

 

[rafaelsmith]

Somehow this thread has devolved into snake oil...whats next...we need to put a battery in our HDMI cables?...maybe lift them off the ground some?

 

[nello]

Getting, hopefully, back to the @Rmarci thread, I have a number of questions. @Rmarci has said in the past he mostly tests HDR versions because that is what he is interested in. The leads me to the following questions:

1. I see that K has recently released updates to a number of HDR versions, all/many of which @Rmarci has pointed out defects in. K has not, as far as I can see, updated the DV versions of those movies. Can I assume, the since the HDR and DV versions are done through separate encoding, that the DV versions don't suffer from the same defects as the HDR versions?
2. @Rmarci - have you ever tested both the HDR and DV version of a movie and saw that they don't share the same set of defects?
3. Has anyone spot checked the DV encodes to see if they also suffer similar types of defects?

I know that K has released updates to some DV movies. Maybe the HDR only updates are a timing thing and the DV versions are coming. I would be disappointed if the DV versions suffer the same defects but are not being updated (i.e. the squeaky wheel gets the oil and @Rmarci is pointing out the HDR versions so that is what K is fixing).

 

[Rmarci]

This thread is the right place to discuss this kind of thing, but please at least provide scientific measurements. What you see or hear with your eyes or ears is very subjective and not really helpful.

I never claimed to know anything about how the HDMI signal is processed internally, and I don’t care about semantics. I only care about objective input/output measurements of the HDMI signal. Based on the hundreds of lossless captures I’ve done, I see no indication that the player is increasing noise.

Maybe once the firmware finally fixes the unintended DNR and block artifacts, some noise will show up and if that happens, I’ll be the first to report it with actual data. And btw, the Ugoos does introduce some noise in the signal, i talked about it a while ago when i was testing if my AVR affected the signal...

 

[Rmarci]

Getting, hopefully, back to the @Rmarci thread, I have a number of questions. @Rmarci has said in the past he mostly tests HDR versions because that is what he is interested in. The leads me to the following questions:

1. I see that K has recently released updates to a number of HDR versions, all/many of which @Rmarci has pointed out defects in. K has not, as far as I can see, updated the DV versions of those movies. Can I assume, the since the HDR and DV versions are done through separate encoding, that the DV versions don't suffer from the same defects as the HDR versions?
2. @Rmarci - have you ever tested both the HDR and DV version of a movie and saw that they don't share the same set of defects?
3. Has anyone spot checked the DV encodes to see if they also suffer similar types of defects?

I know that K has released updates to some DV movies. Maybe the HDR only updates are a timing thing and the DV versions are coming. I would be disappointed if the DV versions suffer the same defects but are not being updated (i.e. the squeaky wheel gets the oil and @Rmarci is pointing out the HDR versions so that is what K is fixing).

1- Yes, HDR10, SDR, and DV P5 each get their own encode, so I assume no issue was found in the DV version.

2- Unfortunately, I cannot test the DV encode properly because the player’s conversion to HDR10 lifts the absolute black pixels (64).
Using my Vertex to process the LLDV signal fixes this lifted black issue, but then other problems appear because the RPU (dynamic metadata) is used for tone mapping according to the EDID target brightness. Using an EDID target higher than the content MDL, for example 10,000 nits, which avoids any tone mapping but causes additional issues, and the KS player seems to suffer from the same overbrightening bugs as the other CMV 2.9 players (Oppo, Sony, etc.).

This is strange because the KS is supposed to be a CMV 4.0 player, and none of the other CMV 4.0 players I’ve tested had this overbrightening bug when using an EDID target higher than the MDL. So either the KS is actually only CMV 2.9, or Dolby was implemented the same way as on the Ugoos, where only TV-LED RGB tunnel supports CMV 4.0 while LLDV is limited to CMV 2.9. In that case, the Ugoos will show the same overbrightening bug in LLDV, but a clever workaround has been implemented by competent coders: they limit the EDID target to the RPU MDL. The ugoos also fix the cmv2.9 LLDV crushed black (80) issues.

 

[thrawn997]

1- Yes, HDR10, SDR, and DV P5 each get their own encode, so I assume no issue was found in the DV version.

Are the block artifacts the firmware bug or encoding issue? I need to double check but I thought I was watching DV when I saw those

 

[Rmarci]

Are the block artifacts the firmware bug or encoding issue? I need to double check but I thought I was watching DV when I saw those

Firmware bug: they are visible in all three modes: SDR, HDR10, and DV.
The recently updated titles were about something else: color accuracy and a fake grain overlay.

 

[thrawn997]

Firmware bug: they are visible in all three modes: SDR, HDR10, and DV.
The recently updated titles were about something else: color accuracy and a fake grain overlay.

Ah, I am a bit less sensitive to noticing the color accuracy differences, the block artifacts are very distracting. I thought the fake grain overlay was due to the firmware activating upscaling, didn't realize they were adding that into their encodes! Strange they would do that for HDR10 and not DV.

 

[Rmarci]

didn't realize they were adding that into their encodes! Strange they would do that for HDR10 and not DV.

They definitely do, because I’ve shown some old vs. new encodes comparison that clearly demonstrated the grain overlay was gone, Prometheus and Kingsman, IIRC.