or Connect
AVS › AVS Forum › Video Components › Video Download Services & Hardware › Amazon Prime Instant Video
New Posts  All Forums:Forum Nav:

Amazon Prime Instant Video - Page 29

post #841 of 1235
Personally it's difficult for me to believe that Amazon is entirely responsible for writing and maintaining Instant Video players for the hundreds of devices which now have them. If they were, I'd expect them to take the same tact that Netflix, Hulu Plus and YouTube have taken and produce a single player UI design with a device-specific layer which they leave up to the OEMs to port (based on an example reference implementation) and a portable UI layer requiring little or no modification from platform-to-platform. I have Netflix players on Roku 2, PS3, Xbox 360, Panasonic DMP-BDT220, -BDT110 and Sony BDP-S390 BD players and my TiVo Premiere and all have identical Netflix players except the Roku 2 and Xbox 360 (the Roku 2's player has the same "flavor" as the common one; all Xbox 360 streaming players are unique inasmuch as they're given MS' "Metro" look and feel and augmented for voice and gestural control via Kinect). Their Hulu Plus players are also identical, except for the Xbox 360's.

There are Amazon Instant Video players on all of my aforementioned devices other than the TiVo (which has a non-Prime-eligible downloaded Amazon video player); every single device's Amazon player is unique, though they share some design elements here and there. IMO, that makes it unlikely that Amazon has a lot to do with development of them.
post #842 of 1235

After doing my testing on the 2009 Panasonic device, and then logging into my account on my PC, I discovered something interesting:

 

http://www.avsforum.com/t/1415137/amazon-instant-video-vs-netflix-quality/90#post_22402982

post #843 of 1235
Quote:
Originally Posted by michaeltscott View Post

Personally it's difficult for me to believe that Amazon is entirely responsible for writing and maintaining Instant Video players for the hundreds of devices which now have them. If they were, I'd expect them to take the same tact that Netflix, Hulu Plus and YouTube have taken and produce a single player UI design with a device-specific layer which they leave up to the OEMs to port (based on an example reference implementation) and a portable UI layer requiring little or no modification from platform-to-platform..

Whether it's Amazon, Netflix, or any other, the player makers have no control over what application is fed to their player. In the case of Amazon, the list of supported devices is pretty small relatively speaking and Amazon actively resists adding more to the list. I'd suspect that they don't want unlimited users accessing their servers. And, they further want to control the bandwidth used by those who do - hence they limit access to 5.1 audio.

Further, the different applications used by Amazon or Netflix all have very different requirements for processing power in the player and more importantly they have different capabilities with respect to DRM. Point being, there are many unseen reasons for different applications being available to different devices.
post #844 of 1235
Quote:
Originally Posted by rdgrimes View Post

Whether it's Amazon, Netflix, or any other, the player makers have no control over what application is fed to their player.
We'll have to agree to disagree about that; as I stated, I doubt that Amazon created any of the players and I'm certain that some Netflix players were written by their manufacturers.
Quote:
In the case of Amazon, the list of supported devices is pretty small relatively speaking and Amazon actively resists adding more to the list.
It's still hundreds of separate models, including devices from just about every major CE OEM: Panasonic, Sony, Samsung, LG, Microsoft, TiVo, Roku, Vizio, Netgear, Logitech, etc. I cannot believe that they keep a firmware development team large enough to maintain unique players on all of those platforms. The list of devices has grown steadily--what evidence do you have of them "actively resisting" adding more? The only mobile devices that they list players for are models of the Kindle Fire line (though I know there's one in the iTunes store) but that's just to promote their own products in that space.
Quote:
I'd suspect that they don't want unlimited users accessing their servers. And, they further want to control the bandwidth used by those who do - hence they limit access to 5.1 audio.
Amazon's HD w/DD5.1 sound streams average 2.3 Mbps including sound; the SD stereo ones are 1.3 Mbps. We're not talking mountains of bandwidth here.
Quote:
Further, the different applications used by Amazon or Netflix all have very different requirements for processing power in the player and more importantly they have different capabilities with respect to DRM.
As I stated, the vast majority of newer devices with Netflix players all run the same player and Amazon's highest bandwidth content is half the bit rate of Netflix's. It's counter intuitive that an Amazon player would require more resources than one for Netflix. All those devices run identical Hulu Plus players (as nice as any of the Amazon ones) and the VUDU players on those which have them are all the same (save for the Xbox, with its Metro look-and-feel and Kinect control features). I don't buy your arguments.
post #845 of 1235
Just noticed the following Trek movies are now Prime:

TMP
II Wrath of Khan
IV Voyage Home
V Final Frontier
VI Undiscovered Country
First Contact
Insurrection

Not included yet:

III Search for Spock
Generations
Nemesis
2009
post #846 of 1235
Quote:
Originally Posted by michaeltscott View Post

Amazon's HD w/DD5.1 sound streams average 2.3 Mbps including sound; the SD stereo ones are 1.3 Mbps. We're not talking mountains of bandwidth here.

I thought so as well.

I hadn't used Amazon for some time partly for this reason and mainly because most of what I wanted to see on Amazon Prime was also available on Netflix (at1080P).

However last night I tried both Iron Man 2 and some Dr Who Season 6. Both claiming HD and 5.1

I was surprised to see the average download speed was ~3 Mbps and picture was better than I recall - still not as good as a good Netflix 1080P stream - but not bad.

I used 10 minute averages to get the 3 Mbps - initial download was much higher presumably as it filled a buffer. The 3Mbps was after the initial peak was no longer in the 10 minute average - so presumably the buffer was full and download speed indicative of the actual video stream.

EDIT Playback device was ROKU 2 XS on a 20 Mbps Comcast connection.
post #847 of 1235
For Netflix, data consumed in 10 minutes will be 10 minutes worth of content plus the unplayed buffer, so the true average has to consider that. (You can establish the size of the buffer by playing something for a while and then pulling the network cable and timing it with a stopwatch until it runs out of buffer and stops playing; I just did that with my Roku 2 XS playing Amazon and got an astonishing 4 minutes; the typical Netflix player buffer--including Roku 2's--is about 90 seconds long).

My Roku is perpetually in debug mode and when playing an HD title with 5.1 sound it says, "Playing 2.5 mbps stream on your xx.x mbps network". (Actually it says the same thing when playing an HD titles which don't have 5.1 sound so it's obviously not a precise statement.

I just measured minutes 5 through 14 of Ong Bak 2 (I use it because it has a high-action, challenging-to-encode sequence of martial arts in the rain near the beginning and because it's available from Netflix, Amazon, Zune and VUDU so I could do sort of apples-to-apples comparisons). Below you can see the bandwidth consumption graph of that same video sequence as played on the Roku 2's Amazon player (on top; no 5.1 sound in the Amazon version) and as played on the Roku 2's Netflix player (the bottom, 1080p video w/5.1 DD+ sound):

Graphs (Click to show)
Roku2XSAmazonHDOngBak2Small.jpg

Avg - 2674.69 kbit/s, Peak - 13997.00 kbit/s, Total - 191.31 MB


Roku2720p20OngBak2.jpg

Avg - 4752.05 kbit/s, Peak - 17208.31 kbit/s, Total - 339.89 MB

The Amazon player seems to top off its buffer by downloading data at high speed, then playing completely out of the buffer for several seconds.
Edited by michaeltscott - 10/13/12 at 10:06pm
post #848 of 1235
Quote:
Originally Posted by michaeltscott View Post

For Netflix, data consumed in 10 minutes will be 10 minutes worth of content plus the unplayed buffer, so the true average has to consider that. (You can establish the size of the buffer by playing something for a while and then pulling the network cable and timing it with a stopwatch until it runs out of buffer and stops playing; I just did that with my Roku 2 XS playing Amazon and got an astonishing 4 minutes; the typical Netflix player buffer--including Roku 2's--is about 90 seconds long).
My Roku is perpetually in debug mode and when playing an HD title with 5.1 sound it says, "Playing 2.5 mbps stream on your xx.x mbps network". (Actually it says the same thing when playing an HD titles which don't have 5.1 sound so it's obviously not a precise statement.
I just measured minutes 5 through 14 of Ong Bak 2 (I use it because it has a high-action, challenging-to-encode sequence of martial arts in the rain near the beginning and because it's available from Netflix, Amazon, Zune and VUDU so I could do sort of apples-to-apples comparisons). Below you can see the bandwidth consumption graph of that same video sequence as played on the Roku 2's Amazon player (on top; no 5.1 sound in the Amazon version) and as played on the Roku 2's Netflix player (the bottom, 1080p video w/5.1 DD+ sound): Warning: Spoiler! (Click to show)
Roku2XSAmazonHDOngBak2Small.jpg
Roku2XSNetflix1080pOngBak2Small.jpg
The Amazon player seems to top off its buffer by downloading data at high speed, then playing completely out of the buffer for several seconds.

Michael - for those of us with older eyes (or less-capacious screens), it would be nice to put at least the most important summary numbers in your text! (And you can change the "Warning: Spoiler!" tag to your own choice of label - that's fully-editable text when you're setting it up.)

Substantively, could you run this same test with something that is in 5.1 on both systems? Otherwise it's even more apples vs. oranges than it has to be, given that it's already comparing 1080p to 720p!

-Phil
Edited by Philnick - 10/14/12 at 12:57am
post #849 of 1235
Quote:
Originally Posted by Philnick View Post

Michael - for those of us with older eyes, it would be nice to put at least the most important summary numbers in your text! (And you can change the "Warning: Spoiler!" tag to your own choice of label - that's fully-editable text when you're setting it up.)
Substantively, could you run this same test with something that is in 5.1 on both systems? Otherwise it's even more apples vs. oranges than it has to be, given that it's already comparing 1080p to 720p!
-Phil

Sorry--I wasn't really offering those two graphs for serious comparison, just to illustrate how different the buffering schemes of the Roku 2 Amazon and Netflix players are. I've replaced the 1080p/5.1 Netflix graph with a 720p/stereo one (I was surprised that I had that at the ready). How's that biggrin.gif?

I'll try to remember to dig the relevant numbers off the graphs; I'm viewing this stuff on a 46" LCD panel, so reading the numbers on those screen captures isn't too challenging.

I was completely unaware that the spoiler tag in this otherwise crappy board software had a text label parameter; I'm pretty sure that vBulletin's spoiler tag did not. I'm old and I never use the little controls on this "editor" to insert mark-up or I'd have seen that. The "Warning!" label is what you get by default if you don't specify anything. Thanks for pointing it out.
post #850 of 1235
Quote:
Originally Posted by michaeltscott View Post

For Netflix, data consumed in 10 minutes will be 10 minutes worth of content plus the unplayed buffer, so the true average has to consider that. (You can establish the size of the buffer by playing something for a while and then pulling the network cable and timing it with a stopwatch until it runs out of buffer and stops playing; I just did that with my Roku 2 XS playing Amazon and got an astonishing 4 minutes; the typical Netflix player buffer--including Roku 2's--is about 90 seconds long).
My Roku is perpetually in debug mode and when playing an HD title with 5.1 sound it says, "Playing 2.5 mbps stream on your xx.x mbps network". (Actually it says the same thing when playing an HD titles which don't have 5.1 sound so it's obviously not a precise statement.
I just measured minutes 5 through 14 of Ong Bak 2 (I use it because it has a high-action, challenging-to-encode sequence of martial arts in the rain near the beginning and because it's available from Netflix, Amazon, Zune and VUDU so I could do sort of apples-to-apples comparisons). Below you can see the bandwidth consumption graph of that same video sequence as played on the Roku 2's Amazon player (on top; no 5.1 sound in the Amazon version) and as played on the Roku 2's Netflix player (the bottom, 1080p video w/5.1 DD+ sound): Graphs (Click to show)
Roku2XSAmazonHDOngBak2Small.jpg
Avg - 2674.69 kbit/s, Peak - 13997.00 kbit/s, Total - 191.31 MB
Roku2720p20OngBak2.jpg
Avg - 4752.05 kbit/s, Peak - 17208.31 kbit/s, Total - 339.89 MB
The Amazon player seems to top off its buffer by downloading data at high speed, then playing completely out of the buffer for several seconds.

Yes the Amazon player fills its buffer and then plays from it - I see the same.

That's why I let it fill up its buffer (at up to 27 Mbps) at the start of playback - and then let it run for 15 minutes or so - at that point the 10 minute average should be a good estimate (the buffer was full and so the download should indicate the actual audio/video stream bandwidth).

On both Iron Man 2 and Dr Who Season 6 the average over 10 minutes (when the buffer was already full) was 3 Mbps - I will post charts when I get a chance.
Edited by undecided - 10/14/12 at 12:42am
post #851 of 1235
See Dr Who from around 15 - 25 minutes on Closing Time from Season 6 on Amazon Prime



and Iron Man 2 on Amazon Prime

post #852 of 1235
Quote:
Originally Posted by undecided View Post

Yes the Amazon player fills its buffer and then plays from it - I see the same.
That's why I let it fill up its buffer (at up to 27 Mbps) at the start of playback - and then let it run for 15 minutes or so - at that point the 10 minute average should be a good estimate (the buffer was full and so the download should indicate the actual audio/video stream bandwidth).
On both Iron Man 2 and Dr Who Season 6 the average over 10 minutes (when the buffer was already full) was 3 Mbps - I will post charts when I get a chance.

I still think that 10 minutes of raw data represents 10 minutes played plus the stuff in the buffer. If we assume that it's 840 seconds worth of content (14 minutes), the 226.51 MB consumed in your top graph represents a 2.3 Mbps average (the "MB" is 2-to-the-20th bytes) and the 221.88 MB in your bottom graph is an average of 2.26 Mbps. (Tomato expresses the total in 2-to-the-20th bytes, which is forgivable, but also express the average rate in 2-to-the-10th bits, which isn't rolleyes.gif).

Now I have to complain--you've shrunk your graphs even more than I do and I can just barely read the numbers even from close up to this 46" screen. Is the average on your top graph 3192.84 kbit/s and the bottom 3102.12 kbit/s? (Yeah--it works out).
post #853 of 1235
Quote:
Originally Posted by michaeltscott View Post

I still think that 10 minutes of raw data represents 10 minutes played plus the stuff in the buffer. If we assume that it's 840 seconds worth of content (14 minutes), the 226.51 MB consumed in your top graph represents a 2.3 Mbps average (the "MB" is 2-to-the-20th bytes) and the 221.88 MB in your bottom graph is an average of 2.26 Mbps. (Tomato expresses the total in 2-to-the-20th bytes, which is forgivable, but also express the average rate in 2-to-the-10th bits, which isn't rolleyes.gif).
Now I have to complain--you've shrunk your graphs even more than I do and I can just barely read the numbers even from close up to this 46" screen. Is the average on your top graph 3192.84 kbit/s and the bottom 3102.12 kbit/s? (Yeah--it works out).

I don't think it is filling the buffer in these 10 minutes - at the start of each show I can see a few minutes of download at much higher speed presumably filling the buffer (the average for the first 10 minutes of both shows was well over 4 Mbps). Once the buffer is full what we'll see represents the actual video data rate.

Anyway it all conjecture - but Amazon Prime does look better than I remember.

It is still not as good as good Netflix 1080P streams and certainly not HDX (we watched Promethius last night on Vudu HDX and it was stunning) - but Amazon Prime is OK for content not available elsewhere. Frankly that is mainly the reason I don't use it as most of what I could see on Amazon Prime is available on Netflix already.

(Oh I am a little surprised about the legibility comments - on my 24" screen and aging eyes I can read the numbers smile.gif )
post #854 of 1235
I like Amazon Prime because I've had that subscription for years anyway for the expedited shipping, so there's no incremental cost for me. If I really like a film, I'll usually end up buying it anyway, so I use it as a free taste-test, as opposed to renting the disk, which adds to the cost if I decide to buy later.

I don't subscribe to Netflix because my older Roku XDS - which I need for its optical output for surround because my amp has no HDMI input - is stuck at 720/stereo for Netflix, so Amazon is actually better for me, given that there's a fair amount of 5.1 material on Prime that I can get with my setup. (It can do 1080p, but about the only service that sends that to it is the 1080P Showcase, which is scraped from YouTube.) My Oppo BDP-93 gets YouTube and Vudu in 1080, but it also is stuck with a 720/stereo Netflix app!

-Phil
post #855 of 1235
Quote:
Originally Posted by undecided View Post

I don't think it is filling the buffer in these 10 minutes - at the start of each show I can see a few minutes of download at much higher speed presumably filling the buffer (the average for the first 10 minutes of both shows was well over 4 Mbps). Once the buffer is full what we'll see represents the actual video data rate.
We'll have to agree to disagree about that smile.gif.
Quote:
...Amazon Prime does look better than I remember.
It is still not as good as good Netflix 1080P streams and certainly not HDX (we watched Promethius last night on Vudu HDX and it was stunning) - but Amazon Prime is OK for content not available elsewhere. Frankly that is mainly the reason I don't use it as most of what I could see on Amazon Prime is available on Netflix already.
It's pretty good; I've been watching Fringe over from the beginning; the entire first four seasons are available as Prime Instant Video in HD with 5.1 sound; it's not available on Netflix at all.

I using forum member RangerOne's TVandMoviesNow.com search engine because when something available both as a Netflix stream and as Prime Instant Video, it will sometime have better rendition on one than the other. For instance, Iron Man 2 is available in HD from both services, but is available with 5.1 sound from Amazon, but not from Netflix (it's available with closed captions from Netflix but not from Amazon, if you prefer). On Netflix, Iron Man 2 is one of those relatively-recent-release titles which don't have a 1080p encode, so video should be comparable; I'd probable prefer to watch it on Amazon with superior sound, which an action movie is improved by.
post #856 of 1235
I have noticed that 'The Last Airbender' is correct aspect ratio and 5.1 on Amazon and cropped and stereo on Netflix - so maybe I do need to check out Amazon a little more.


TVandMoviesNow.com looks useful - thanks.
post #857 of 1235
Quote:
Originally Posted by michaeltscott View Post

We'll have to agree to disagree about that smile.gif.
It's pretty good; I've been watching Fringe over from the beginning; the entire first four seasons are available as Prime Instant Video in HD with 5.1 sound; it's not available on Netflix at all.
I using forum member RangerOne's TVandMoviesNow.com search engine because when something available both as a Netflix stream and as Prime Instant Video, it will sometime have better rendition on one than the other. For instance, Iron Man 2 is available in HD from both services, but is available with 5.1 sound from Amazon, but not from Netflix (it's available with closed captions from Netflix but not from Amazon, if you prefer). On Netflix, Iron Man 2 is one of those relatively-recent-release titles which don't have a 1080p encode, so video should be comparable; I'd probable prefer to watch it on Amazon with superior sound, which an action movie is improved by.

I watched Iron Man 2 on Amazon a week or so ago and it was quite good. I did find that the LFE channel was a bit heavy and had to trim it back in my AVR - but that's a "problem of abundance" - the best kind of problem to have!

-Phil
post #858 of 1235
Where does the title page have the info if it's 5.1 or not? I'm trying to find out if my TV, Panasonic UT50, supports Amazon's 5.1 but I'm not sure where to look for the audio format. Right now my receiver says PCM, but I expect that regardless of how many channels it's receiving.
post #859 of 1235
Quote:
Originally Posted by anikun07 View Post

Where does the title page have the info if it's 5.1 or not? I'm trying to find out if my TV, Panasonic UT50, supports Amazon's 5.1 but I'm not sure where to look for the audio format. Right now my receiver says PCM, but I expect that regardless of how many channels it's receiving.

When you click through to the screen where you have the choice whether to play it, add it to your Watch List, etc, it should have the format information down in the lower left hand corner of the screen. It'll say 5.1 there if it's in 5.1. I assume that if your player doesn't support it, it won't say that.

As far as known quantities go, check out anything in the firefly tv series and the movie that came out of it, Serenity; season one of the original Star Trek tv series, the wonderfully quirky Raising Dasies, Iron Man 2, Captain America, the First Avenger, Thor.

If your player supports 5.1 and your connection is fast enough, those should play in 5.1 for you.

As far as I know, however, very few devices get 5.1 from Amazon - the easiest ways to get 5.1 from Amazon are with a Roku or a PlayStation.
Edited by Philnick - 10/26/12 at 4:07pm
post #860 of 1235
Quote:
Originally Posted by anikun07 View Post

Where does the title page have the info if it's 5.1 or not? I'm trying to find out if my TV, Panasonic UT50, supports Amazon's 5.1

It doesn't.
post #861 of 1235
You can check out a list of movies and television programs available as Amazon Prime Instant Video with 5.1 sound here, at TVandMoviesNow.com. It may not be entirely up to date; the site's owner, AVS Forum member RangerOne, tells me that he has to update his list of Amazon content via a long manual process and only does it occasionally.
Quote:
Originally Posted by Philnick View Post

As far as I know, however, very few devices get 5.1 from Amazon - the easiest ways to get 5.1 from Amazon are with a Roku or a PlayStation.

As you say, you can play Amazon Instant Video with 5.1 sound Playstation 3 and Roku, but also on some LG, Samsung, Sony and Vizio televisions and STBs. Not Xbox, not Panasonic stuff.
Edited by michaeltscott - 10/26/12 at 5:32pm
post #862 of 1235
Thanks guys. I didn't really expect the TV to stream 5.1, but I was hoping. My S580 BDP is supposed to support it so I can test it there, too.
post #863 of 1235
Quote:
Originally Posted by anikun07 View Post

Thanks guys. I didn't really expect the TV to stream 5.1, but I was hoping. My S580 BDP is supposed to support it so I can test it there, too.

The BDP-S590, -S390 and -S790 do.
post #864 of 1235
Someone in the Roku forums pointed out this news about Amazon now giving people an $8/month-cancel-anytime option for Prime membership. Just like Netflix's streaming-only plan with a currently smaller library, but with no-additional-charge 2 day shipping on items sold or fulfilled by Amazon and access to the Kindle Owner's Lending Library (a couple hundred thousand books which can be read for free on Kindle devices, one per month). Nice move; I wonder how many new Prime members this will get them? I'm sure that there are those who are put off by the $80 lump-sum yearly payment who'd be willing to commit $8/month.
post #865 of 1235
post #866 of 1235
post #867 of 1235
Has anyone ask the question if OPPO is going to offer Amazon Prime?
I guess I just did.
Thanks
post #868 of 1235
Quote:
Originally Posted by st johng View Post

Has anyone ask the question if OPPO is going to offer Amazon Prime?
I guess I just did.
Thanks

The answer is no. Amazon has refused to offer it to Oppo.
post #869 of 1235
Quote:
Originally Posted by rdgrimes View Post

The answer is no. Amazon has refused to offer it to Oppo.
Or are they just asking for too much in licensing fees for what is basically a rather small company?
post #870 of 1235
Quote:
Originally Posted by rdgrimes View Post

The answer is no. Amazon has refused to offer it to Oppo.

I always thought the idea of the ap was to get it into as many devices as possible, thus selling more monthly plans ..
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Video Components › Video Download Services & Hardware › Amazon Prime Instant Video