1725 posts • joined 26 May 2011
Re: or maybe
Maybe a little harsh there but I sympathize. This seems pretty harmless as it will be trained pilots within sight and restricted to sets. Personal usage is becoming a huge issue here. I have no issues with people wanting to fly them over public land (or their own land) assuming they can do so safely but sadly some morons think it's ok to fly over private property and film. If I sit in a tree behind your house and took pictures of your kids in the garden I'd likely get arrested for voyeurism or something similar, how is it any different to do it with the camera on a drone? Luckily our HOA have for once managed to do something positive and dealt with the last two drones flying over, I haven't seen the old men so agitated since the great leaky taps disaster of 2011). There's no public land or roads within half a mile yet some nonce wants to film on HD (last one looked like a 5d mk3 with a 16-35 on it) low enough to be within range of a garden hose? I just don't get how people think it's ok.
I think we need some firmer guidelines add to what's ok and what's not. As long as we are sensible we can probably allow more commercial use but the amateurs seriously need some restrictions.
How about banning all tablets from stadiums. Even better remove all rear facing camera from tablets so inconsiderate twonks don't block everyone's view because they just have to record an entire event rather than just watching it.
Re: We don't even have 4G yet!
Yeah there was a lot of fluffing going on with 4g. We do have 100 mbps. A 20x20 MHz FDD LTE deployment exceeds this on paper but that is shared and assumes perfect signal (in theory you might get 140mbps) etc but on a lightly loaded network fils are seeing 80 mbps.
Using LTE A or cat 6 or Rev 9 or whatever name you want to use :) will allow bonding of channels but in all honestly that's marketing fluff for gaining the highest speed tests. Thera not much you need to do with a phone that uses more than 4mbps. The overall capacity isn't increasing with carrier aggregation. It's just you can access multiple carriers at once which is actually not ideal add the lower spectrum carriers should be left for people who can only access those. You would be unimpressed if you were near the edge of a cell, out of range of higher frequency carriers and the lower frequency was crowded with chumps closer to the antenna who are just phallus stroking over the higher aggregate speed.
Mimo is already ruled out and working here, extra capacity and better quality at the edge of the range which is great, that's probably the next thing you'll see along with CA. Some mid 2000-3000MHz will probably come along to boost capacity as well.
I'm very interested in the ability to operate full duplex that they created last year. That blows the Shannon limit out of the water. That had to be on their list of tricks for 5g.
I realize it is country dependant but 100 mbps even when mobile is a reality for quite a few cities here. Even in the sticks I can get 30 mbps here and that'd on a 5x5 carrier.
+1 for Waitrose. I always found their prices to be very close to Asda and Tesco but crucially their vegetables lasted longer. I got sick of stuff like broccoli only lasting a day or two before discolouring from Chavco. Waitrose had an own brand vegetable line that comes from their own farms, I guess it just cuts out some of the process so it's quicker to the shelf.
Some of their stuff is more expensive and if you want to live like Patsy and Eddie then sure it's going to cost you but I managed to feed my family from there for the same money as the other places. My only real indulgence was their fresh dijon mayonnaise :)
Re: Flaw in the argument
Couldn't agree more re healthy eating. Plus that twat Oliver had a lot to answer for. Everything that used to be cheap but decent he featured and the price went stupid. Cod cheeks, scrag of lamb (perfect for a crock pot curry), oxtail etc as soon as he mentions it the price triples.
At least horse carrots are still cheap, only decent thing the EU ever did.
Re: Anonymous Coward and Revisionism
Wow, a post deleted by a moderator. I don't think I've seen that before here, to the sites credit they are very liberal with their moderation. The post in question was literally bonkers.
I'd you want to blame anyone for Israel and similar situations a good place to start would actually be my home country, the UK and also France who let uncivil servants sit down with a map and a ruler and carve up large sections of the world with little regard for the people that lived there, their history, religion, language, or tribe.
TBH I wasn't coming at it from a left or right angle, just an honest observation. The US has been a global police force and military force for quite some time. There are some links between the arms industry and politicians that give cause for concern. What confuses me, perhaps worries is a better word, is that there isn't really a consistency. Take Somalia and Rwanda for example.
I try and believe the motives are genuine, most peoples probably are, but often the implementation is sub-optimal. Perhaps it's worse that other countries are slow to move if they do at all when it comes to international military action?
The USA is hardly shy in using force either.
Re: Spot on, ElReg.
I have a note 3 with a zero lemon 10k battery which fits fine in most pockets. Just avoid skinny jeans.
Re: Right upto the point where the Netflix exec demonstrated on camera
Then they would need something akin to DPI to track it, which would be expensive to do with that volume of traffic, it would be cheaper to simply expand capacity. They could simply be targetting IP ranges or hostnames as well, although it's easy enough to switch with a dynamic setup like streaming. Rather than saying they are or aren't doing it, I'm simply trying to suggest ways of narrowing down the variables to find out if they are. I'm discussing the method not the outcome.
DPI tends to work where there is a value added element, such as providing virus scanning for a large network of hosted machines or in a targeted environment, such as with a specific target IP looking for types of traffic. To do it at a backbone level would be expensive. That isn't to say Comcast et al couldn't target netflix traffic, just that the method they would use would likely just be based on rate limiting certain ports or ip's. It would certainly be interesting to find out, but the situation is a little more complicated than simply, its faster via a proxy ergo they are rate limiting.
Re: income will then come directly from consumers
Actually, yeah let's have free transit! I'll cancel my cable account and get a 1 gbps drop for free please!
This I agree with totally. I would be interested to see how cable companies reacted to Netflix if they weren't also competitors.
Re: income will then come directly from consumers
Entirely directly rather than directly and indirectly :) these days transit costs are mostly to do with install costs and the cost of maintaining enough capacity. Whilst you do pay per mbps the cost is actually very low. This is also guaranteed capacity, you home connection isn't guaranteed, it's a shared connection. You may have your own link to the dslam or similar but after that is a contended service. It's likely you will get your full speed most of the time but just like the phone system (pots) I'd everyone tried to use it at once it would fall over. If you needed guaranteed phone access you could pay for it but the average consumer didn't get that, nor did they pay the same price. Contended service equals lower cost.
When you buy transit you get guaranteed access to the full capacity or a partial refund. You also get support 24 7, if there's a router dropping packets somewhere you call and a real tech reroutes it while it gets fixed. This is well worth the low fees paid for transit. You can also blend cheaper and more expensive providers to get a more cost effective mix.
Say Skype can me out with a 3rd holographic Web calling service that used 20 mbps connections and it took off. If you didn't use it and just used your connection for say YouTube and downloading iso's how would you feel if the cost of your service went up to pay for additional backbone capacity that want related to what you use the Internet for. Transit fees just help place the cost where is accrued and are very different to charging an extra fee for priority status for traffic already on a network. NN is all about providers wanting to charge extra for data they have already been paid (with either currency or services in kind) to carry via intermediaries.
Re: Right upto the point where the Netflix exec demonstrated on camera
They aren't using DPI, they would just use ports to identify types of traffic, so an unencrypted proxy wouldn't change anything. A proxy running on the same port Netflix sends video on would be interesting.
Re: And what about...
So you expect they will just make less money? Seriously you couldn't pour water out of a boot if the instructions were written on the heel. Look at the amount of detailed explanations offered to you and look at what you have offered. You obviously have no experience of how the industry works or the technical aspects at play. You spout opinions dressing them up as fact and you expect to be able to change the industry. I know the industry, you obviously don't.
Good luck with that :)
Re: Seems you've already bought into the ISPs arguments...
Google also built a global backbone of their own so they could offload much of their traffic at free public peering exchanges (note not actually totally free, you have to accept data in return and pay port&rack fees fees). I find it interesting that Netflix didn't attempt this as it would have allowed them to differentiate themselves from similar offerings (redbox, amazon prime etc). They obviously have a big market lead and a great product but amazon could do them some damage with prime instant video.
Re: And what about...
Its also worth considering the amount of money in play here. Transit costs vary wildly based on physical location, commit level and carrier. Having 10mbps hooked up to your cabin in the woods in the middle of nowhere from a tier one provider will cost more per mbps than a 10gbps drop in a carrier hotel \ exchange etc.
Netflix is talking about transit in well served areas and in very large amounts. These days it's not uncommon to pay between $600 and $2000 a month for a 1gbps drop and $4500 to $9000 for a 10gbps (ipv6 is actually starting to affect pricing as well). If you say a person streams at 3mbps for an average of 4 hours per day the actual cost to Netflix is around 30 cents a month, maybe a little more. This is hardly stifling innovation, it's a couple of percent of the monthly charge.
The NN debate started when mobile phone companies started to talk about charging extra for skype etc on their networks, traffic that was already being carried. They were looking at a way of protecting their voice income and also subsidizing lte build outs without having to raise consumer pricing. The fixed line industry wasn't on the radar until Netflix started running up against congestion. Note they didn't get traffic shaped, they simply hit the limits of their bandwidth providers and started to get politiky. Sadly people bought the BS. I love Netflix as a service but the game they are playing is just outright dishonest. You don't hear their story from any other company, they create over 30% of peak time traffic in the entire country, they are just being asked to pay fees that cover the port costs and some of the back end infrastructure to support that, the same as every other company with servers does directly and indirectly and hasn't complained about it for the past 40 years.
If you want to legislate how their business model works you need to do it via a PUC which is fine, I mean it doesn't work as our water and electric companies routinely prove but thats the manner. If you push for legislation over a situation you don't understand you aren't going to get it. If you do, it won't do what you want it to.
I get that it seems nonsensical to pay at both ends, just take a moment to consider the other view. This is not anti NN, this is simply about sharing the costs where they are incurred. Rather than placing the entire cost of the connection on one end it is shared between both. Time Warner et al are simply not just going to accept less revenue if one source is removed. They will just transfer it or use it as an argument that they need to be able to charge for QOS exceptions. The very thing you actually want to avoid. It isn't a simple situation, but all the information is out there, the history with the wireless companies etc.
Re: And what about...
nice try. So explain, you cut off this source of revenue, do you think ISP's will just make less money or that they will charge end users more? Nobody ever answers this :)
Most educational institutions in Europe pay to connect to an educational network which then connects at peering points. Much of their traffic is internal so free peering is sufficient. They do already pay. There is so much misinformation from people on this, it's sad, if you care about the situation take the time to understand it.
Re: And what about...
erm that would be an intranet,the internet is a bunch of connected networks. They have two customers, both pay. Yes they make profit but what makes you think they wont just charge end users more if you abolish transit fees? You are missing the point and you are going to end up hurting yourself.
ok one last try :)
Please try and understand, there is a difference between paying (directly or indirectly) for access to a network and paying a premium for priority on a network. (FWIW I am ok with the former but not the latter)
Net Neutrality is about last mile providers charging a premium for priority for data already on their network.It's also interesting to consider how this plays alongside QOS and network management. You naturally want to prioratise time senstive activities like VOIP, Streaming (radio and music), 'Skypeing' etc over say email or downloads. This does improve a customers experience where there is congestion. The issue is if they exclude video from that because its a competing service or if they were to allow priority for a video service that paid a premium. Note paying for transit is not a premium, it's a simple reality of providing something, anything, over the internet.
There is a real issue with Net Neutrality and I understand its easy to misunderstand but if you do support NN then please try and take the time to understand how the system works. Misunderstanding it just allows the anti NN lot to win and the real problem is allowed to occur, which is the paying for a priority. Netflix was never asked to pay for a priority, they were advised that they needed more capacity. None of the ISP's were placing any priority on transit ingress over peering ingress. Netflix sadly seems to be trying to muddy the water.
Re: On paper OK, but
It will be interesting to see how this plays out. If transit fees were abolished then monthly sub rates would go up and companies like L3 would disappear which would result in a degradation of service and an increase in costs. L3 et al are more than just aggregators, they also keep pricing down as they can drop a chunk of their traffic via free public peering and via cheaper networks like HE or Cogent (which is probably posh by modern standards) to dilute their costs. This forces tier 1 isps to keep their prices vaguely reasonable. It's worth noting that Netflix is paying less to the tier 1 isps than it did to level 3, oh those mean isps!!
Sadly it would appear the cant think wont think crowd with no concept of the situation have locked onto this. At least its amusing to watch. It's like listening to my cat debate string theory :)
Re: On paper OK, but
Yes CDN's still exist and buy transit (and use free peering) from last mile networks, so they do exactly what Netflix is having to do. CDN's basically offer an off the shelf alternative to building out your own global network and delivery system, which are two separate things. If you want to build your own servers but not have to negotiate with every major ISP you use an intermediary company who has a backbone connected to major ISP's. They also will have some level of peering and as they aren't a tier 1 ISP they will also be paying for transit from Tier 1 ISP's. Basically they are brokers and aggregators who do some of the delivery for you but then subcontract the rest. The point being with both these models you still pay directly or indirectly to the last mile network as long as their customers.
I do think you misunderstand the financial issue here. You state they don't need to worry about the amount of bandwidth, they do, they pay for it. It's not like they just pay to upload a 1GB file and that's it. They then pay for the number of times it's downloaded which goes to pay the ISP's for the last mile access. Essentially this is what happened. Netflix used an intermediary company to 'remove' their data, this companies links to several ISP's became saturated. The intermediary didn't want to pay for more links, the last mile ISP refused to give it for free so individual connection speeds slowed down.
>The ISP's already charge each and every customer for that bandwidth.
I already explained above, they charge at both ends of each network, they run a last mile network which is paid for by the people accessing it and they run a backbone paid for by companies accessing it. Please read my post before assuming I am wrong.
>But then the ISP(s) started getting greedy. And now they still want their customers to pay them, but they also want the companies that their customers are using to pay them too.
Absolutely false. Charging for transit from companies connecting servers to the internet has been around for decades. It is not something new, the backbones have been funded by transit fees since commercial backbones were released. The cost of operating the networks is shared between customers at both ends based upon usage levels. If you want to pay for a server in your building you pay a one off connection fee and a monthly fee based on the capacity of the pipe you get or 95th percentile billing for people stuck in 1980.
>But then the ISP quietly starts slowing down the data flow for Netflix
Again, completely false. Time Warner etc certainly could have slowed down connections using routing rules \ QOS rules but they did not, they didn't need to. The company Netflix paid (Level 3 were the main one iirc) for its connection to the internet had links to the last mile networks. These links were based on a commercial agreement, either a peered link where you agree to a free transfer of data assuming the ratio one way to the other doesnt exceed X:1 (X is usually between 3 and 5), or transit where they buy a link for $x a month for Y mbps. All that happened was Level 3 sold routes to Netflix at a certain rate, the egress points for their network became congested and this resulted in individual streams having to drop to a lower rate as they were sharing the same finite connection. Level 3 could have expanded their agreement with the last mile isp's and simply paid more. Level 3 were actually charging Netflix more per mbps than TimeWarner wanted for a direct connection.
> Netflix in the end, is forced to approach the big, greedy ISP(s) and offer them money in exchange for what should have been equal treatment by the ISP in the first place since their mutual customer has paid both of them.
Again, you completely misunderstand what happened. Netflix wasn't targeted. If they offered a service that used 1/100th of the bandwidth they wouldn't have had any issues, but given the volume of data they generate they have found themselves needing to directly buy transit rather than indirectly as they previously did. The 'evil' company in this case is level 3, they sold something they couldn't deliver.
I'm not saying net neutrality isn't important, I believe strongly in it, but this isn't a net neutrality issue. Netflix's traffic wasn't artificially limited. They weren't forced to do anything any other company doesn't have to do, they just ran into a limit quicker due to the volume of data they produce. Sadly it is misinformation and not taking the time to understand this situation which hurts net neutrality. Net neutrality isn't about companies paying nothing for transit, it is about prioritizing data based on who you pay.
>That's why Netflix offer a peering arrangement for free
Netflix also offered to generously provide caching servers to large ISP's. If they accepted this arrangement or free peering who pays for the backbone? Caching servers would have been quite sensible but I don't think netflix was prepared to provide them in the numbers required as it wouldn't be one per network, to lessen the load on the backbone it would be one per neighborhood. A consumers monthly charge doesn't reflect the cost of operating a cable network and a backbone. If ISP's offer totally free transit and then pass on that extra cost to the subscribers your monthly bill is going to go up. I'm not sure that is a win for us as consumers. Then anyone with a server gets to connect to the internet as much as they want for free and pass on the cost to us even if we don't use their service.
As I said above, Net Neutrality is about not prioritizing data, not it being free. Companies expect to pay for bandwidth, go and check out large server providers like Softlayer, they pay for their connections to ISP's (without complaint) and their customers pay for it. Bandwidth is fairly cheap, usually cheaper than the power costs or amortization of the hardware unless you are doing something data heavy like sending out millions of 2-4mbps streams. What is at stake is can an ISP charge extra to give priority to data already on its network? That is what we want to avoid.
There is a different between charging a company more for transit because they compete with you and simply charging for transit. If we abolish charging for transit (which may be the only way to prevent abuse) that income will then come directly from consumers pockets rather than indirectly.
Whilst charging for transit is legitimate in principal and had been used fairly in the past I don't trust companies that also sell TV services to act fairly when considering whether to peer or sell transit to the likes of Netflix.
A large isp like time warner effectively operates two networks. A last mile network to you home which joins a second network which is a national backbone. In theory when you buy a cable connection you are paying for your share which is the last mile network. The cost of the backbone comes from companies buying transit to hook up servers to the Internet. Isp's also peer which is basically a free exchange of data, this usually had a maximum ratio to stop it becoming too unbalanced and unfair. Where this gets messy is they also offer peering to important sites, or more accurately free transit. The big question is how ok is it to offer free transit to say amazon.com but then charge Netflix. Then how do we regulate that. I know in advertising one tv station will charge a premium to a competitor, I.e. What sky charges virgin for slots, so is it fair for one industry to charge a competition premium but not another?
It's not a simple cable companies are evil situation, they sure are, but charging for transit is not really evil. Netflix isn't really a net neutrality issue, they aren't being asked to pay extra for priority, they are just being asked to pay for access like many companies have to do. They had exceeded the capacity of the intermediary companies links and had a choice between augmenting those links or just outright buying everything directly. The party at fault was the intermediary company which sold transit they couldn't handle. Personally I'm not sure which way to lean on this as a whole because I think the cable cos would have peered with Netflix if they hadn't been a competitor, but charging for transit is fair.
Re: I repeat my comment from the last article on this subject:
I would like to see any legal action resulting in the removal of the executives who made the decisions rather than just a fine which will likely result in front line staff (who weren't at fault for the breach) being fired to 'make economies'.
Net Neutrality? Hold my coffee, I got this: FCC says it's still considering all options for Open Web
Re: <rolls eyes>
Depending on the stage at which it is added to the bill it can actually remain anonymous, so not only can they do crap like this (or at least something similar) they can do it without being found out.
Re: Good cables are better
Do your Thunderbird cables have gold plated connectors?
Re: ok, but seriously...
The kardashian thing is mostly media manufactured. Actual famous people have gotten better at hiding some of not most of their lives and the media have started creating all these pseudo celebrities who are famous just because they act like idiots where they can be witnessed. They are happy for the attention because they are vapid, meaningless dregs incapable of doing anything worthwhile with their lives, the media love them because other vapid plebs pay to watch them and plebs love them because they act like they wished they could act of only they didn't need to have the door widened to leave their house.
It's absolutely sickening but basically it's the great unwashed, can't think won't think crowd driving this. The media print what sells and the scroats get off on this crap.
Sad, but that's the world we live in.
Thank you! Far too often this is overlooked which is ridiculous.
Interesting, did I read it right? They are using 3 separate chips for the PA's. Qualcomm's gobi PA's have 2g, 3g, HSPA and LTE all on one chip. I would have thought that would better suit apples simplicity approach better, perhaps the power use or unit cost was higher? Are the non qualcomm chips GAAS or CMOS?
Re: Time to move on
Yes and no. Their sensors are improving which is great. Their actual camera range and lenses is a nightmare. Everyone they get an ecosystem remotely established they have invented a new mount. They also seem to cost more than canon or nikon (compare the 70-200 2.8 is/vr lenses, canon is cheaper). So to recap they have the a mount range in both ff and apsc, the nex mount and now the FE mount. There is some cross compatibility but it usually stuffs performance royally. They have managed to release some very tempting bodies but the lense range just isn't there. Going mirror less was smart because their slt (inspired heavily by canon) approach exaggerated their sensors biggest flaw at the time, poor low light performance. The issue was they had gaps in their previous lens lineup, they were trying to flesh out the apsc slt line and refresh the aging full frame lineup their inherited from Minolta. They had to start over with nex lenses and now again with FE. Its also not as simple as just extending the mount to make up the different flange distance as you need to move the rear element for optimal performance. If they wanted to do it they needed to throw more resources at it and price stuff to sell. You can't price your 300 2.8 higher than canon and expect to sell many. Canon and Nikons latest are leg humpingly good. The Sony is close but doesn't beat them. Some of the Zeiss glass is amazing but again priced high. Weather sealing was another deal breaker.
They will get there but it's a shame is taking so long.
Re: The Horror Of Marketing-As-Management
As a photographer, looking from the outside in, Kodak ' s biggest mistake seemed to be the half hearted approach to digital and the belief that going all in on dslrs would kill their film business. Sadly they seemed to forget that other companies might make dslrs. They made some great products, besides film (I have to admit to preferring fuji film) there were a couple of impressive dslrs ( the14n/c were epic) and some great ccd sensors that went into medium formats. If they had developed their own body and mount along with a range of lenses to replace lost film revenue I think they would still be around and a big player.
So rather than infighting and the film division restricting digital development it was just plain bad management? Sad, I really liked kodak.
Re: Sony is a great example of what happens when you implement...
Did they? I know the world went 6 sigma crazy at one point (I was a tangerine belt project ninja champion warlord) but besides a weekend rafting and a few extra meetings it was fairly harmless. I thought they just missed the shifting of some core markets like TV and computing by being too stubborn and not investing enough in keeping ahead, although I am probably totally wrong.
I'm not saying you are wrong, just curious about you thoughts on what they did.
Re: In before
Is he still around?
Perhaps this is karmic justice for inflicting trinitron rage upon the world. For those who haven't owned a trinitron set, this rage occurs when you get your squillion pound (in both weight and money) trinitron set home and bask in its awesome quality then some cnut points out the two little wires that you were happily oblivious to before. Then all you can see forever more (as those sets are actually indestructible) is those two little +×÷=ing wires across the middle of the screen. It was even worse with the trinitron monitors. The upside was the pigs were well fed.
Re: Well I'm still waiting for an appollogy
Canon design and fab their own sensors, at least in all their dslrs.
Nikon do use Sony sensors in their low and mid range cameras but their top tier camera sensor is designed in house and fabbed by someone else, usually Renesas. This may change in the future. Nikon don't tend to just directly use Sony sensors, the optical side of the sensor is pure Sony but the digital conversion uses a chunk of their own IP which is why you see the same resolution but better noise control even in RAW on a nikon using a Sony sensor.
Hassie, phase one, pentax, and Olympus have started to use Sony sensors although not exclusively in some cases. Iirc fuji had them fab their sensors but it's a fuji design. But no, canon is all in house using lithography from 1920 which isn't far away from using felt tips on cling film over a lamp shade.
Sad news indeed, Sony have made some great products over the years (I had a trinitron set, the first 3 play stations and we have a Sony flat screen and bluray now) but also some stinkers, I nearly got one if their phones but iirc it had a fixed battery. Their camera and sensor business is doing very well. It would be sad to see the great parts of the company sink because they held onto poorly performing areas.
Re: Samsung's Business model
I read it (and as I understand it from elsewhere) as the 14 billion covers all their products wheres the 6.9 is just the profit from one area of business.
I would love to see what happened if they spent the entire marketing budget on r&d.
I'm sorry to hear that. Is this something you have been able to either demonstrate on a larger sample size or fully explain the mechanism for?
Journal etc do tend to limit work from outside academia because it's more likely for an academic to follow the required protocol and present a robust proof. I'm not attempting to belittle your work, quite the opposite. Firstly I'm glad you are recovering. Secondly, to take it further you need to try and repeat your results (which to be fair can be tough) on a wider sample size, do so in a controlled manner (so it's ethical, safe for the subjects and you are better able to prove that the factor you are controlling is the driving the change) and hopefully also demonstrate how it actually works.
It may be worthwhile approaching people working in the field to work with you and share credit.
Re: psychology ~= science...
This is true, to be a science there must be explosions (or implosions), maths gets an exception for its ability to make people's heads explode. All the rest is just basket weaving.
Re: Children of the Night...
Ditto. The peak at night is likely due to factors such as concealment or targets being asleep rather than people who are awake late are murderers and perverts. Now if you had said elected officials or employees of a large public service broadcaster then maybe you would have been onto something. ..
You forgot to add inside with the heating on :)
Just curious about the 1 GB RAM. It doesn't make much sense to compare it directly to Android, but in comparison to other iPhone wouldn't the full hd screen require more RAM. The amount of RAM reserved for graphics went up in at least the note series when they went from 720 to 1080, it Apple kept the RAM the same and increased the resolution wouldn't that decrease the RAM left for everything else ? Has anyone seen any impact from this?
Again, genuine question, not an anti Apple rant.
Very true, any concerns I have with it are to do with the fact it doesn't go far enough. I was just having a little fun with the whole Fox blames X on Obama care thing :)
I'm sure this is just to save money. It's cheaper to let China find vulnerabilities in the system then hire a specialist. We have to pay for 'bamacare somehow ;) *cough*
What are your predictions?
Not what you want, but what you think will happen? If there's no exit polls we should just invent our own misinformation based fun :)
I think 55% yes.
Why won't that evil old git just do a Maxwell.
Re: I can feel the difference...
It's complicated and the physics put me to sleep but the extra information does affect the image even if you can't discern the individual pixels. In photography weer use the term micro contrast (probably incorrectly :) ) but having spent years making prints and testing lenses that out resolve either the film or sensor in my eyes the effect is there. Others may disagree, that's fine. Discussing Nyquist frequency and aliasing puts anyone to sleep well before they can learn the math but consider this.
You have a black line one pixel thick on a white background with a single red pixel in the middle. You start close and you can see the line and the red pixel. As you move back and reach the limit of your eyes to discern the indivisible pixels you'll notice they don't actually just vanish like the last slice of pizza at a works party. The distinct red pixel starts to merge with the black making a combination of the two and the black line merges with the white background making a grey stripe. The pixels still affect your vision beyond that point for a period then they do vanish.
At the end of the day you watch a UHD display and look at the price tag and make an individual choice which will be right for you but El Jobs and his retina math was just marketing. Its not right to say it's totally pointless at this level for everyone, that's is impossible to see a difference. Perhaps 16 or 32 know will be part the point which there's any discernable difference for anyone.
Re: Chicken and Egg
Sorry. Just to correct an error, Netflix just licensed the technology from Eyeio, they didn't acquire them.
I just went and had a look at UHD from YouTube. Encoded with h264 at 20 mbps using a mediocre encoder. Viewing it at 1080 showed pretty good quality. Viewing a 100% crop changed it a little. There was definite blocking visible in areas of motion but areas without much motion were crisp. Check out "road to machu pichu" in 4k. It does really show the weaknesses in this type of compression, hevc should negate some of it but not all. The part where the guy is in the canoe, he's rocking so moving but not by much so there isn't much blocking yet the background is moving and more contrasty and therefore showing blocking. If you throw enough encoding power and a decent encoder at it you can reduce the blocking but it's still going to be there with h264 and to a lesser degree with h265. It will be far less noticeable with h265.
I just found this page if you are interested in seeing h265 / uhd at those bitrates.
Not the greatest examples, it's hard to tell with the shutter speed used if the blurring on the moving flowers is due to motion blur or compression artifacts but it looks like the former. No blocking on the moving detail either which is seriously impressive.
Just caught the comment about the bitrates being low compared to physical media. Yup, I wouldn't argue that. Although bluray and streaming use a better algorithm than DVD so that isn't directly comparable. Streaming is at a lower rate and a lower quality, each person can make their own choice if it's worth the trade off. Personally I can live with it for TV stuff and movies on a smart phone but I respect that others wouldn't. I still get movies I like on bluray but personally I do like what Netflix et al offer.
Re: Chicken and Egg
Well for a start, if we are being technical it is UHD, not 4k :) iirc these days 480 streams at about 700kbpd, 720 streams up to 2500 kbps and 1080 up to about 4000 kbps (at amazon, Netflix is higher). Assuming they don't bump up bit depth (which for streaming would be pointless) and assuming a realistic compression gain of 25% to 30% is perfectly reasonable to expect UHD streams at 15000 kbps to exceed the quality of 1080. I will qualify that with the fact that I haven't yet seen their UHD in action so this is based on simple math and the initial offering will probably get better over time as I will explain below.
When Netflix started streaming their quality was very poor, simply because they took an off the shelf encoder and tried to compress the video more than was reasonable. What they did was go and buy a company (eyeio ?) that was writing a much better encoder and that resulted in a huge increase in quality for the same bitrate. I agree totally it is not up to bluray quality levels, there is a tradeoff and judging by the popularity of Netflix, Amazon prime and Hulu it is hard to argue that people aren't prepared to accept that tradeoff for part of their viewing. HEVC is pretty cool in how it works, h264 was good but this takes it to another level. The ability to use coding tree blocks to breaks up what was previously called macroblocks into varying sizes depending on the variation in content is huge. It can use data where is needed most in any individual frame so if you had an area of largely similar colors and detail and an area of disparate color and detail the encoder doesn't have to compromise, it can retain the higher level of detail in one area without having to record excessive detail in the area that doesn't have it. Couple that with the fact that increasing resolution in the same image tends to result in a less than equal increase in bitrate for the same perceived quality, what they are proposing is not unrealistic, especially once the encoding algorithms catch up. I think Netflix were taking about having a 25 mbps upper ceiling on UHD streams which is even more head room for both quality and increased gamut / bit depth.
Before writing it off its probably best to wait and see. I'm not buying into 4k for Netflix or Amazon, but I do think the initial signs are what they plan on is reasonable (if the bar is set at offering a comparable quality to 1080 but at 4x the resolution ). They aren't going to match physical media but that hasn't stopped them from doing well so far.
- +Comment Anti-Facebook Ello: Here's why we're still in beta. SPAMGASM!
- Vid+Pics Microsoft WINDOWS 10: Seven ATE Nine. Or Eight did really
- Analysis Windows 10: One for the suits, right Microsoft? Or so one THOUGHT
- Xbox hackers snared US ARMY APACHE GUNSHIP ware - Feds
- George Clooney, WikiLeaks' lawyer wife hand out burner phones to wedding guests