A comparison of PS3 and 360 as media players

Incorrect. When audio upsampling was first introduced into the firmware, it only worked on disc based sources. This wasn't compatible with XMB multi-tasking either, but since most people play back music from the HDD while browsing, they wouldn't have noticed before.

I hate saying this, but I find people's comments about audio upscaling a bit naive. We know that the DVD upscaling is pretty good, many people claim they can't even tell the difference between DVD upscaling and BD playback. Now while I think that's overstating it, it's clear that the image quality can be improved through upscaling. If you look at sound-waves, wouldn't you think that it could in fact be just as easy or even easier to upscale this? If anything can be improved in quality by 'adding bits' surely a lot of music can be enhanced a fair bit by upscaling.

For me personally, initially I was struck by the difference in quality between playing back an original CD, and playing back the version I copied to the HDD. At some point I realised this was because back then the upsampling wasn't applied to HDD sources. Even now I can clearly hear the difference, and I'd be surprised if most people on this forum can't with even a modestly decent sound system or headphones.

I also believe that if you have a 5.1 surround set, then even if it only supports 44.1 inputs, the upscaling to 5.1 means it can theoretically upscale the audio resolution by a factor of 3 (6 channels versus 2).

It´s a BUG, i disabled any kind of "upscaling" and 2.0 to 5.1 conversion. It should just play it at 44.1 khz and not spend time and energy on creating some weird 48khz conversion that most likely uses some primitive conversion technique.

It used to play 44.1 khz just fine, unless they suddenly apply some kind of magic Mojo to 44.1 khz as well that requires bundles of resources, it´s a bug.

I agree with the upscale part, it was first when oversampling was introduced that CD´s started to sound like they were marketed. It should sound better with the high samplerate and 24bit upsampling.
 
It´s a BUG, i disabled any kind of "upscaling" and 2.0 to 5.1 conversion. It should just play it at 44.1 khz and not spend time and energy on creating some weird 48khz conversion that most likely uses some primitive conversion technique.

I would guess it's a unified output system, that outputs sounds from the XMB, music your playing, sounds in the webbrowser, sounds from voice/video chat, and so on and so forth. All these will be mixed together into 48khz (2.0, 5.1, whatever) for output.

I think I may get your point. Theoretically, if you deselect all but the 2.0 44.1 (and 48, these two can't be deselected) options AND you deselect the bitmapping (this is the actual upscaling method, I just figured out), then it should definitely allow the multitasking. I can see why they may have disabled the option for simplicity (i.e. not have to think about which of the many combinations of outputs and upscaling methods are available), but in theory at the very least when you only selected the 2.0 output options at their base 44/48 sampling rates, then they should have supported this option. May still have posed problems though for playback of video streams from the webbrowser, for instance.

As for the actual upscaling, I found more information here:

http://forum.ascendacoustics.com/showthread.php?t=3433

Now. Onto CD's where this setting actually means "upsampling."

Whenever you resample an audio or video signal (especially with downsampling) you want to dither the signal to spread (and hopefully reduce) overall error. This is obviously needed with downsampling where you are inherently losing information. BUT, with upsampling, dithering can help reduce artifacts, distortion, and noise.

What do they do?

Type 1: Simple dithering. The result will reduce artifacts and potentially help distortion. It will, however, inherently add white noise to the signal increasing the noise floor. This can be helped, which is where Type 2 comes in.

Type 2: Noise shaping. This technique dithers the signal, but also quantizes error in a feedback loop, which is used to direct the dithering. The final step is to push the error to audio frequencies less perceptible by the human ear. The overall error will be the same as dithering, but the error will be less noticeable (to humans). This is how you can have a 24-bit remaster on CD. Noise shaping can effectively maintain the dynamic range and noise of a 24-bit signal for the 1kHz-4kHz range in a 16-bit signal.

Type 3: "Use a technique developed specifically for the PS3 system to enhance audio playback." Obviously a bit of marketing there. It may have been produced by the PS3 team within Sony, but it's just a dithering technique performed on an audio signal, which would work just the same for any device working with digital signals. I'm not sure what technique this is. Just as Type 2 is a more directed and "smarter" extension of Type 1, I'd assume that Type 3 either adds to what Type 2 does or perhaps is a different approach, but still based on Type 1 dithering.

So, Type 2 is certainly better than Type 1. Also, it is probably safe to assume that Type 3 is better than Type 2. If you are worried about the unknown, stick with Type 2.

See the second post.
 
I would guess it's a unified output system, that outputs sounds from the XMB, music your playing, sounds in the webbrowser, sounds from voice/video chat, and so on and so forth. All these will be mixed together into 48khz (2.0, 5.1, whatever) for output.

I think I may get your point. Theoretically, if you deselect all but the 2.0 44.1 (and 48, these two can't be deselected) options AND you deselect the bitmapping (this is the actual upscaling method, I just figured out), then it should definitely allow the multitasking. I can see why they may have disabled the option for simplicity (i.e. not have to think about which of the many combinations of outputs and upscaling methods are available), but in theory at the very least when you only selected the 2.0 output options at their base 44/48 sampling rates, then they should have supported this option. May still have posed problems though for playback of video streams from the webbrowser, for instance.

As for the actual upscaling, I found more information here:

http://forum.ascendacoustics.com/showthread.php?t=3433



See the second post.

Great find, with a new Reciever incoming i will also get to benefit from samples above 88khz :)
 
I hate saying this, but I find people's comments about audio upscaling a bit naive. We know that the DVD upscaling is pretty good, many people claim they can't even tell the difference between DVD upscaling and BD playback. Now while I think that's overstating it, it's clear that the image quality can be improved through upscaling.

Yes, those people need either a better TV or glasses if they cannot tell the difference between DVD and BD which is like night and day. ;)

If you look at sound-waves, wouldn't you think that it could in fact be just as easy or even easier to upscale this? If anything can be improved in quality by 'adding bits' surely a lot of music can be enhanced a fair bit by upscaling.

I think audio and video "upscaling" are different kettles of fish. I honestly don't see how you could enhance a 2 channel 44.1KHz signal. If you interpolate (ramp it up to 48KHz) you are basically putting information there that doesn't necessarily have to match reality (information theory limitations and all) and the 5.1 conversion I see as just any other DSP effect, a cosmetic change if you will.

So, does it actually sound better or is just your perception, i.e. it sounds nicer? Not that there's anything wrong with that.
 
I think audio and video "upscaling" are different kettles of fish. I honestly don't see how you could enhance a 2 channel 44.1KHz signal. If you interpolate (ramp it up to 48KHz) you are basically putting information there that doesn't necessarily have to match reality (information theory limitations and all) and the 5.1 conversion I see as just any other DSP effect, a cosmetic change if you will.

So, does it actually sound better or is just your perception, i.e. it sounds nicer? Not that there's anything wrong with that.

Of course you can enhance a 44.1 khz signal, 48khz may be to little and to akward a resolution to actual get any worthwhile gains, but 44.1 to 88.2 and 16 to 24 bit really adds something usefull.
 
I think audio and video "upscaling" are different kettles of fish. I honestly don't see how you could enhance a 2 channel 44.1KHz signal. If you interpolate (ramp it up to 48KHz) you are basically putting information there that doesn't necessarily have to match reality...
Which is the same as video upscaling...

The upside to audio upscaling is the signal is far simpler than video and we can work within a more confined space of perception and interpretation.

So, does it actually sound better or is just your perception, i.e. it sounds nicer?
I really don't get what you mean by this line! "Sound better" and "perceived as sounding nicer" are exactly the same thing! All sound is a perception. If you meant "does it rebuild a more accurate representation of the analogue source, or does it just make the errors in the digital source less obvious" than upsampling is a bit of both. The nature of the sounds being upscampled means data in the frequencies that can't be reconstructed can't be heard or wouldn't be wanted anyway. Maybe there was a spike 1/88,000th of a second in length (which is unnatural, so wouldn't be sampled anyway unless there was an electric storm around or something) that wasn't captured in a 44kHz sample, but that doesn't mean anything to the human ear which can't perceive it, or wouldn't want a pop of a rogue bit either. So a signal reconstruction that misses that spike in a smooth wave wouldn't be missing any relevant information that'd affect the user's experience. But also the purpose isn't to rebuild exactly the source wave, but to create a more pleasing experience. Take an old 256 colour VGA display. Dithering wasn't at all accurate in creating original artworks or smooth gradients, but the end result was far superior to having ugly bands. Audio dithering does exactly the same. I suppose this is only possible though because the dithering is at a higher frequency than human perception. Dithering an 8 kHz signal in upscaling it to 16 kHz would be adding a very high but audible frequency with a constant, very low amplitude 8 kHz sine wave (if I've got this right).
 
Can someone tell me if I should change my PS3 RGB output settings to Limited instead of Full, when watching videos (eg DVDs, Blurays, video files on the HDD)?

As I understand only games support the Full RGB spectrum?
 
Can someone tell me if I should change my PS3 RGB output settings to Limited instead of Full, when watching videos (eg DVDs, Blurays, video files on the HDD)?

As I understand only games support the Full RGB spectrum?

If your TV supports RGB Full settings, then you should set both settings to full (also the super white thing). The PS3 makes sure that whatever doesn't support Full RGB as a source, is mapped to the Full RGB spectrum anyway.

Think of it like this: say I have a color range of 0-10, where 0 is no color, and 10 is the full red. If the PS3 outputs this range, it would map this range either to 16-235 or to 0-255 (Full RGB). Now if I have a source color range of 12.000 to 512 million, it would still map this to either 16-235 or 0-255, depending on your setting. In this case, the source material doesn't matter as much as you'd think, see?

However, this is still the best setting if your TV supports it, because 0-255 has 36 more steps of color information per pixel. So whenever you output information that uses 255 steps, you get the best color resolution, while you lose nothing (with room for better color upscaling in theory when you have stuff like DVD upscaling active on the PS3) whenever source material that is 16-235 is mapped to the 0-255 of Full RGB by the TV.

At least that's how I understand it. I just read the thread patsu posted, but people there don't actually fully understand what's going on here I think, except a few people at the end of the thread. :p
 
Last edited by a moderator:
Thanks, my Bravia definitely supports Full RGB. I was just wondering whether the PS3 adjusts a source using the limited RGB spectrum to use the Full RGB spectrum without introducing artifacting or diluting blacks etc. You've answered that, so I'l set it permanently to Full RGB.
 
NPD now thinks that Blu-ray will enter mainstream sometime in 2010:
http://www.blu-ray.com/news/?id=3474

NPD uses five adopter categories: “cutting-edge” consumers, “sensibles” (older people who can afford new technology but don't rush to buy it), “dreamers” (those who want new gadgets but are deterred by price), “neutrals” (people who are unimpressed by technology) and “anti-tech” (those who believe new gadgets are a waste of money).

The most encouraging sign for major Blu-ray acceptance was the progress within the two more mainstream groups: sensibles and dreamers. They are considered vital to a product's sales because they make up about 42% of the U.S. population. In February 2008, they made up just a small fraction of the Blu-ray usage base, at 22% combined. By August 2009, that had increased to 37%. In other words, out of every 100 owners of a Blu-ray player, 37 were “sensibles” or “dreamers.”

The largest group within the BD installed base is still that of cutting-edge consumers, although its predominance has decreased from 64% in February 2008 to 38% in August 2009. Cutting-edge consumers make up just 20% of the US population, not enough to move products into the mainstream.

Blu-ray usage has even increased among the “neutrals” and “anti-tech”, who now make up 24% of the BD user base, up from 14% in February 2008.

There are also rumors that Apple will introduce Blu-ray iMac fall this year:
http://www.blu-ray.com/news/?id=3473

I hope it's a burner instead of a read-only player drive.

Anyway, with the Blu-ray market heating up, hope Sony will improve it some more. Managed Copy and 3D movie have been mentioned so far. Some BD-Live advancement would be good too.
 
NPD uses five adopter categories: “cutting-edge” consumers, “sensibles” (older people who can afford new technology but don't rush to buy it), “dreamers” (those who want new gadgets but are deterred by price), “neutrals” (people who are unimpressed by technology) and “anti-tech” (those who believe new gadgets are a waste of money).
:oops: So I'm a "sensible" "anti-tech" "dreamer". I'm so conflicted! :runaway:
 
Amongst all things, I am MOST keen on a Blu-ray burner. Have too many Windows VM (32 Gb each) to archive. I can't keep them reliably on HDD because someone may use them accidentally, and it's a waste to use my active HDD space to archive them.

Managed copy and BD-Live (Blu-ray App Store anyone ? ^_^) are the other focus areas for me.
 
According to BDA, Blu-ray is indeed outpacing DVD growth in the same lifecycle:
http://www.avforums.com/forums/home...live-blu-ray-disc-assoc-press-conference.html

By the 3rd year of introduction, BD is outperforming the growth of DVD. Yes, BD is growing faster than DVD, Color TV, VHS, etc. DVD is often seen as the fastest growing CE product - not any more.

That is down to the Playstation 3. It is a true statement from the BDA, but I thought I'd add this caveat just to explain any surprise.
 
That is down to the Playstation 3. It is a true statement from the BDA, but I thought I'd add this caveat just to explain any surprise.

Not sure if that makes sense? The PS2 helped DVD too after all, especially in Europe and Asia.

I think a much bigger reason here and one I predicted earlier on, is that you can play your existing DVDs on BluRay players, often with better picture too (upscaling). Now if you need a new DVD player, you'll more and more likely just get a BluRay player because they're not that much more expensive and you're future proofed, etc. It's just a much easier transition than VHS to DVD was. Back then for instance VHS was also recording device, while DVD couldn't yet for a long time. Now people will do mostly harddisk recording, and it doesn't really matter if you pair that up with DVD or BluRay.
 
That is down to the Playstation 3. It is a true statement from the BDA, but I thought I'd add this caveat just to explain any surprise.

Yap, that's why I posted the links in this thread instead of Video Tech.

PS3 plays a big part to shift Blu-ray sales forward. Some execs mentioned that about 74% of PS3 owners watch Blu-ray (obviously at different frequency).

Sales of standalone players have also picked up because of price drop (About US$150 for cheap ones).

Amazon Blu-ray sales and NetFlix low-cost Blu-ray rental help too.

The stats was collected before PS3 Slim introduction. So we should see another jump forward, especially if the iMac rumor turns out to be true.



The other thing for the studios to work on is digital download. I know Sony has been secretly/quietly testing day-and-date releases for digital distribution and disc-based releases. iTunes, NetFlix, XBL Video Marketplace, and Playstation Video Store will certainly benefit from the new digital download business models.


All in all, consumers will get to choose how they want their movies: Full ownership and rental model.
 
Back
Top