Can Red Change the Internet?

[M Most]

I have Cox's regular cable internet and right now I am on my old laptop on the other side of my house from the router. I just checked my speed and it was download 8.54 and upload 2.1 so Red Ray should work for many I'd assume.

i think you're confusing megaBits with megaBytes, a very common error. I can rather confidently guarantee you're not getting 8.5 MegaBYTES per second on any cable modem in the US.

 

[Stuart English]

RED RAY at 2K?

RED RAY at 2K?

Is it because of RedRay's proprietary compression that allows you to use less data but still maintain superior 4k image quality?

It is Anthony. The average data rate for RED RAY at 4K resolution is 2MB/s (16Mb/s) ...which is substantially lower than most 1080p Blue Ray titles.

 

[Brian Petrie]

i think you're confusing megaBits with megaBytes, a very common error. I can rather confidently guarantee you're not getting 8.5 MegaBYTES per second on any cable modem in the US.

Haha yeah you're right I thought about that...thanks for the clarification I always mix that stuff up...:-/

 

[Nick D]

My question is... What good is serving live streamed 4K cinema or online content when the end user likely has a Walmart $199 Acer special 24" monitor with 1080p resolution or a 15" laptop with 720p resolution or an iPhone / iPad?

Perhaps, it will be more useful for those with home theaters and 50+" screens if they are willing to actually PAY more for content via a Netflix streaming model, perhaps, but 1080p will be here for a lot longer than I desire, being a fan of "Open Resolution"...

(Simply Agreeing with Peter O's comments)

 

[Wesley Scoggins]

Why does it have to be sustained? Why can't you just let it load for 30 seconds and then start watching?

I already have to wait for about 20 seconds for most of my Blu-Ray discs to finish loading on my Magnavox Blu-Ray player, even longer than that for Criterion titles, and that doesn't even include state-up time.

I don't see why if you were streaming 4K content to your big-screen TV, why waiting a minute or so for it to buffer would be unacceptable for most users. Especially with internet connection speeds getting faster and faster.

With Red Ray's compression technology applied to 1080P or 2K content, I don't see any reason why buffering should ever be needed even for modern broadband speeds much less the broadband speeds of a decade from now when 4K TV's start to become the standard.

 

[Erich Ocean]

The technical challenge with RED RAY is going to be the decompression speed, not bandwidth. Wavelets require a lot of cycles to convert to RGB or YUV for display. People will need dedicated hardware to do it in real time, and the resulting RGB 4K image requires *a lot* of bandwidth to get to a display. If the image needs to be scaled down in anything other than 1/2, 1/4, 1/8 etc. increments, that's even more CPU cycles.

In other words, you're not going to see RED RAY in the home any time soon unless RED starts selling boxes to people, a la Apple TV. We might see it in theaters with dedicated playback boxes, though in that case, I don't know why RED's bandwidth target is so low, especially for audio. I freaking *love* the audio quality on Blu-ray disks. Going back to compressed multi-channel audio ain't going to happen for this consumer.

Also, DCI already uses wavelet compression (JPEG 2000). So all RED RAY brings is even more compressed images for the theater. Is anyone really asking for that? If so, who? If RED RAY is meant for theaters, personally, I would target Blu-ray's bandwidth and 50GB capacity. Studios could definitely ship Blu-rays around without any difficulty, and they might see that as a benefit over shipping hard drives.

 

[Pietro Impagliazzo]

Also, DCI already uses wavelet compression (JPEG 2000). So all RED RAY brings is even more compressed images for the theater. Is anyone really asking for that? If so, who? If RED RAY is meant for theaters, personally, I would target Blu-ray's bandwidth and 50GB capacity. Studios could definitely ship Blu-rays around without any difficulty, and they might see that as a benefit over shipping hard drives.

Erich, does it matter if it's compressed or not if the images are 4K and stunning?

It reminds me of people wanting uncompressed RAW on the RED One... Makes no sense, because the compression scheme is already damn good.

Considering you don't need as much information as RAW for distribution, and considering REDCODE is very efficient, this low bitrate delivers.

At least everyone that saw it projected couldn't believe the bitrate was so low.

I freaking *love* the audio quality on Blu-ray disks. Going back to compressed multi-channel audio ain't going to happen for this consumer.

I'm not particularly familiar with RED RAY audio compression scheme, but yeah, I think we've come to a point where we can give ourselves the luxury of uncompressed audio.

 

[Stephen Gentle]

Also, DCI already uses wavelet compression (JPEG 2000). So all RED RAY brings is even more compressed images for the theater. Is anyone really asking for that?

Videos encoded with H.264 looks a hell of a lot better than MPEG2 videos at twice the bitrate...

'More compressed' means nothing. Look at the picture, not the numbers.

 

[Stuart English]

Audio data rates

Audio data rates

I'm not particularly familiar with RED RAY audio compression scheme, but yeah, I think we've come to a point where we can give ourselves the luxury of uncompressed audio.

So after creating video compression to reduce the bit rate for 4K Digital Cinema quality movies to 15Mbps,

we'd then say "we're done" and turn around and use another 9 Mbps for 7.1 Ch uncompressed audio?


I know audio quality is critical, but that doesn't sound like the RED development team that I work with

 

[Pietro Impagliazzo]

So after creating video compression to reduce the bit rate for 4K Digital Cinema quality movies to 15Mbps,

we'd then say "we're done" and turn around and use another 9 Mbps for 7.1 Ch uncompressed audio?


I know audio quality is critical, but that doesn't sound like the RED development team that I work with

So you're saying that the audio compression scheme within RED RAY is also proprietary?

That would be cool.

 

[Stuart English]

Distribution options

Distribution options

Also, DCI already uses wavelet compression (JPEG 2000). So all RED RAY brings is even more compressed images for the theater. Is anyone really asking for that? If so, who? If RED RAY is meant for theaters, personally, I would target Blu-ray's bandwidth and 50GB capacity. Studios could definitely ship Blu-rays around without any difficulty, and they might see that as a benefit over shipping hard drives.

Agreed. No reason why we couldn't deliver 4K Digital Cinema quality movies to theaters on a 50GB Blu Ray disk.

However, with our compression we can also deliver them on a 16GB flash media card, or via internet download.


And that opens up non-DCI applications for 4K imagery as well as providing DCI distribution another option.

 

[PatrickW]

or via internet download.

RedRayBay.com ???

 

[Erich Ocean]

'More compressed' means nothing.

It means, precisely, that the overall bandwidth requirements are lower. I didn't imply anything else. In particular, I didn't say anything or imply anything about the quality (good or bad) of the decompressed images. I haven't seen RED RAY or done any comparison tests with other encoding strategies on the same footage to render a judgement on that front.

Personally, I'm not sure that movie studios need better compression for DCI footage (or better quality at the same compression). What they have now works fine; bandwidth and storage size aren't issues AFAIK, so I'm not really sure what the market for RED RAY is in theaters.

In the home, the challenge I pointed out above remains: very high CPU/GPU requirements to decompress wavelet-based footage, making it a challenge to decompress footage on existing machines in real time (so consumers can't play it back without special hardware).

 

[Erich Ocean]

So after creating video compression to reduce the bit rate for 4K Digital Cinema quality movies to 15Mbps,

we'd then say "we're done" and turn around and use another 9 Mbps for 7.1 Ch uncompressed audio?

Of course not. I'm actually surprised RED is developing RED RAY at all. Knowing what I know about compression, and how the studios are about standards, I personally wouldn't spend ANY time on a distribution codec for 4K, wavelet-based or not (and I love wavelets, without them I couldn't do my current project).

But I can tell you than on the consumer front, for high end content, the market really doesn't want anything but uncompressed PCM or losslessly-compressed multi-channel audio, preferably at 24-bit, 48Khz or higher resolution. It's easy for non-audiophiles to hear the difference, because the difference is huge. Since the bandwidth requirements are so low (9Mbit -- actually, I've seen peaks at 15Mbit on some discs), people feel comfortable demanding that (and they should). Audio on Blu-ray has been a big win, and a major driver of adoption in the groups I'm in.

If it were me, I'd bump the compression bandwidth for 4K footage up to 35Mbit/sec, removing whatever CPU-cycle killers you could in the process, and keep the existing Blu-ray codecs for multi-channel audio. But like I said, I can't really tell what RED RAY is targeted at, market-wise.

 

[Roger Crouse]

I'd like to log on to redflix.com to stream district 9, the lovely bones, the book of eli, or whatever the next great movie shot on red will be... someday soon... in incredible quality. Red changed the game on the front end - acquisition. Might as well do it on the back end - distribution.

 

[Stuart English]

Audio Data Rates

Audio Data Rates

... on the consumer front, for high end content, the market really doesn't want anything but uncompressed PCM or losslessly-compressed multi-channel audio, preferably at 24-bit, 48Khz or higher resolution. It's easy for non-audiophiles to hear the difference, because the difference is huge. Since the bandwidth requirements are so low (9Mbit -- actually, I've seen peaks at 15Mbit on some discs), people feel comfortable demanding that (and they should). Audio on Blu-ray has been a big win, and a major driver of adoption in the groups I'm in.

I understand your point of view, but a counter-point is the most successful digital audio product ever is Apple's i-tunes.

i.e. heavily compressed.


For us, as 15 Mbit/sec is our 4K resolution video data rate, using the same data rate for audio seems almost perverse.


But we do agree with you about using at least 24-bit, 48KHz sampling to maintain sound fidelity...

... and on a compatibility note, our RED RAY player provides up to 8 channels of uncompressed LPCM audio over HDMI.

 

[Eren Ozkural]

As far as expensive REDRAY compatible consumer media players are concerned there are two factors to think of:

Economies of scale and Moore's law.

RedRocket cards debayer 4K RAW at real time speeds and cost $5000. RED have shipped approx. 8000 cameras as of right now. I don't know what the sales figures are for the rocket but it must be less than that. I'm assuming that most camera owners haven't bought one whereas post houses have bought between 1 and 8. Whatever the figures are, I don't think that they've shipped more than 5-10,000 RedRockets.

10,000 units is but a fraction of the consumer market. If they were to simply shove the guts of a RedRocket into a consumer set top box that plans to ship in the millions the costs would come down considerably.

If you throw away the horsepower needed to debayer the full camera RAW image and just decompress finished footage you can get away with lower powered/cheaper hardware. Add the time needed for consumer adoption of 4K displays and (i'll be VERY generous here) you have 6 years before Sony, Panasonic, Sky et al. would need to consider 4K capable media players for the home. According to Moore's law that's 4 cycles so the processing power/cost ratio would have improved by a factor of 4.

Pure conjecture: If consumer level Red Ray players cost $1,000 to manufacture today the cost could come down as much as $500-$750 in 6 years time. Add discounts for mass manufacturing and you're looking at pocket change per unit.

 

[Jay Gannon]

As far as expensive REDRAY compatible consumer media players are concerned there are two factors to think of:

Economies of scale and Moore's law.

RedRocket cards debayer 4K RAW at real time speeds and cost $5000. RED have shipped approx. 8000 cameras as of right now. I don't know what the sales figures are for the rocket but it must be less than that. I'm assuming that most camera owners haven't bought one whereas post houses have bought between 1 and 8. Whatever the figures are, I don't think that they've shipped more than 5-10,000 RedRockets.

10,000 units is but a fraction of the consumer market. If they were to simply shove the guts of a RedRocket into a consumer set top box that plans to ship in the millions the costs would come down considerably.

RedRocket sales/economy of scale figures must include the card that its derived from. (won't mention as I don't know RED's policy on it.) But suffice it to say there's more than 10,000 out in the wild under different names.
Hardware specific acceleration is always going to be expensive, but RedRay doesnt have to debayer, it just has to decode which is cheaper and easier.

 

[Steven Caesare]

...

In other words, you're not going to see RED RAY in the home any time soon unless RED starts selling boxes to people, a la Apple TV.

...

Red has announced a consumer (ie non-Pro) version.

-sc

 

[Steven Caesare]

...

Since the bandwidth requirements are so low (9Mbit -- actually, I've seen peaks at 15Mbit on some discs), people feel comfortable demanding that (and they should).

"They should"?

Sadly this tends to be based on everything _OTHER_ than objective real-world testing.

I've spoken witha number of folks who claim to be able to hear the difference. remarkably few of them had ever done real double-blind testing to dtermine if they could tell the difference or not. And most of them strangely refuse to do so.

And of those who were willing to partake in the test I conducted, they didn't score any better than what a random guess would be statistically.

I wold much rather have high-fidelity compresseed audio in a smaller package than submit to the placebo effect and have to double my bandwith to get audio along with my 4K video... but then again I've never cared for $1500 "oxygen free long-grain silver" power cables either.

I'm glad Stuart & Co. are looking to bring us high fidelity audio and video in as compact a package as possible.

-sc