(2016-08-24, 20:11)danfloss Wrote: [ -> ]No power on off I'm afraid, apparently something called 'the Uboot' is broken. The version of LibreELEC i have is here https://forum.libreelec.tv/thread-1172.html
I haven't updated to 7.0.2.006 but someone on there said they have
That's enough reason to avoid them for me. IIRC it's because Nexbox aren't arsed to fix the firmware - in android it suspends rather than powering off. There are alternatives that don't have that issue (Beelink for one).
My wifi's rubbish too - not an issue for me but could be a nasty surprise for some.
On android on the Nexbox A5 It will power off buy holding down the power button and will suspend buy just pushing the button... The 95 is a cheaper ver of the A5 and I think they work the same. So far for me the A5 is doing good running everything I have played. No over heating or shutters it been a decent box so far. I have a couple different boxes to compare it 2
(2016-08-28, 09:47)Ned Scott Wrote: [ -> ] (2016-08-28, 09:25)mo123 Wrote: [ -> ]Any one seen this false advertising of 2 Ghz of Amlogic S905, S905X & S912 tv boxes?
http://www.cnx-software.com/2016/08/28/a...dvertised/
Wow, as in, Amlogic themselves were being deceptive about this. That's pretty messed up.
Huge discovery. I wonder how it took so long to notice since so many companies are using these chips. Poor internal testing when these things have to be dicovered by end users. This affects virtually every media box manufacturer using this generation of Amlogic chips since they all advertised the 2 GHz speeds. Processing claims from unhappy buyers will be very expensive.
It is not good to cheat by amlogic but if someone is now happy from his box (this how it works etc etc) why now should be unhappy . Box will work fine. Nothing change. We will see what amlogic now do
(2016-08-28, 15:26)RoarTiger Wrote: [ -> ] (2016-08-28, 09:47)Ned Scott Wrote: [ -> ] (2016-08-28, 09:25)mo123 Wrote: [ -> ]Any one seen this false advertising of 2 Ghz of Amlogic S905, S905X & S912 tv boxes?
http://www.cnx-software.com/2016/08/28/a...dvertised/
Wow, as in, Amlogic themselves were being deceptive about this. That's pretty messed up.
Huge discovery. I wonder how it took so long to notice since so many companies are using these chips. Poor internal testing when these things have to be dicovered by end users. This affects virtually every media box manufacturer using this generation of Amlogic chips since they all advertised the 2 GHz speeds. Processing claims from unhappy buyers will be very expensive.
It's hard to find things like that. Everything tells you that it's working fine so how do you find that it doesn't? Especially when nobody expected this kind of cheating. Look how much effort they needed to prove that there is something wrong.
Anyway it's really bad that there are things like that happening. I wonder what was people at Amlogic thinking. I mean...didn't they know that sooner or later someone will make this discovery and that could mean quite big loss in a future?
(2016-08-28, 15:26)RoarTiger Wrote: [ -> ] (2016-08-28, 09:47)Ned Scott Wrote: [ -> ] (2016-08-28, 09:25)mo123 Wrote: [ -> ]Any one seen this false advertising of 2 Ghz of Amlogic S905, S905X & S912 tv boxes?
http://www.cnx-software.com/2016/08/28/a...dvertised/
Wow, as in, Amlogic themselves were being deceptive about this. That's pretty messed up.
Huge discovery. I wonder how it took so long to notice since so many companies are using these chips. Poor internal testing when these things have to be dicovered by end users. This affects virtually every media box manufacturer using this generation of Amlogic chips since they all advertised the 2 GHz speeds. Processing claims from unhappy buyers will be very expensive.
To be completely honest, it's a meaningless spec. CPU frequency has never been a measurement of speed or performance, unless comparing two CPUs of the same processor family. Listing the CPU frequency is one of the great lies of tech marketing, and it's been going on for decades. It's about as relevant as knowing how much the CPU/SoC weighs. You can easily have a 1GHz CPU outperform a different 2 or 3GHz CPU. Because this affects all S9xx SoCs there was never anything to compare it to.
IMO, the only reason this is an issue is because it's something Amlogic appears to be lying about, to the point of making the amlogic-provided software lie about what frequency is being used. As others have said, if they lied about this then they could be lying about other things.
Some might say that it matters because they wouldn't have bought a product with an SoC that said "up to 1.5 GHz", but that's foolish. Those numbers never represented performance. You can't even compare the S8xx line to the S9xx line using CPU frequency. That being said, it is still wrong for Amlogic to lie about a spec, even if that spec doesn't have a realistic meaning in this situation. It must be very tempting to cook the numbers like that, because most consumers don't understand and think that higher is always better. It still doesn't make it right.
I stopped looking at cpu frequency a long time ago, it honestly doesn't matter in todays tech unless you have no idea about tech.
(2016-08-29, 06:20)Ned Scott Wrote: [ -> ] (2016-08-28, 15:26)RoarTiger Wrote: [ -> ] (2016-08-28, 09:47)Ned Scott Wrote: [ -> ]Wow, as in, Amlogic themselves were being deceptive about this. That's pretty messed up.
Huge discovery. I wonder how it took so long to notice since so many companies are using these chips. Poor internal testing when these things have to be dicovered by end users. This affects virtually every media box manufacturer using this generation of Amlogic chips since they all advertised the 2 GHz speeds. Processing claims from unhappy buyers will be very expensive.
To be completely honest, it's a meaningless spec. ...
I understand well the disconnect between MHz and actual performance but that doesnt affect liability regarding false advertising which was the point of my comment. It is promoted as part of the product, consumers have every right to expect that the item meets the listed specifications. Some of the margins on these boxes are very low; this debacle could wipe out all profits from selling these products. Even a successful company such as your employer may be severely impacted if it has to process claims for all the devices (Hub and Play 2) it sold because they do not meet the advertised speeds. Of course this is all still being investigated so it may turn out that these chips do meet spec in the end and no one will have liability issues, but the preliminary results dont look good.
Of course, and I expect WeTek will have some kind of official response once they have had time to contact Amlogic themselves. All sold Hubs are within their return window, and future products can be clarified as soon as they get notice from AML. I don't think this will impact them significantly, because of the video-focus of the product, but it is right to offer.
I would assume that any losses by the company could then be demanded from Amlogic for reimbursement. However, I am far from an expert on such things.
Like I said before, the real concern isn't about performance. Real world tests were already being done for months and months on the same hardware, and the S905 is an excellent SoC for media boxes as-is. The real concern is, if this is actual/intentional deception from Amlogic, then what else are they lying about. Finding that out is more important to me than achieving some meaningless spec. Dissatisfied customers will be taken care of, but what about the future?
No matter what happens, Amlogic still needs to get their shit together. They could be the king of the TV box mountain, but they need to act like big boys and girls for that to happen. More transparency, following standard APIs, and so on. They're so close, but still so far.
It will be intresting to see what Amlogics official response on this issue is, providing they even publically give a respnse. No doubt the vendors that buy from Amlogic will be asking for answers, whether or not the general end user ever gets a statement who knows.
The average Joe Bloggs end user probably doesnt even care, with only the technically inclined forum users etc actually being aware of the issue. As long as these devices play back the video and audio codecs people need them to with accpetable performance average users probably wont care.Apple for example are always very secretive about the tech specs of their Iphone SOCS in terms of clockspeed, perfomance etc but they still sell by the truckload.
There's also the factor of how much you paid for your device, my Amlogic device is cheap it does what I want it do, my outlay on cash was minimal...
Had I spent the extra cash (double, triple what my device cost me) for a branded premium device I would be a lot more bothered about what I was actually getting.
Considering you expect the companies that charge premiums for their product to have better in house testing facilities so that issues like this can be detected, Hardkernel have come out and said that the 1.75ghz and 2ghz clock speeds don't even exist in the hardware blobs for the AML chips.. Up until now how did something like this slip theough the net ?? Why wasnt it picked up by the extra testing and R&D we expect from companies that charge premiums for their products ? So far fair play to HK they are the only manufacturer that I have seen to acknowledge the issue and are actually trying to address it. Whether or not they go as far as offering refunds to customers for being mis-sold products even though it looks to be Amlogics fault remains to be seen.
The plus point is however though if there is any chance of a end user refund for being mis-sold a product it is much more likely to come from companies with a reputable brand , dont expect to see any rebates on the cheap china devices , but then again you should know this before you buy them.
Amlogic need to be careful , if Rockchip up their game Amlogic may well be in trouble.
Hello
I cannot agree with your dismissal of AMLogic cheating or comparing Soc GHz, and low price does not make fraud legal.
World commerce and trade depends on binding agreed terms.
A Intel and AMD CPU are not the same but everyone expects their CPUs to run at the claimed 2GHz. They expect 2GB of ram to be 2GB, not 1.5 GB.
A Ford, A Vauxhall, A BMW car are not the same but if their Speedometer says you are traveling at 100 miles a hour, you would not accept in truth the car to be doing 50 miles a hour, only.
Fifty dollars is fifity dollars and not 35 dollars and two pebbles.
AMLogic set out to cheat GHz and subroutines to cheat benchmark tests.
Are their Soc 5.1 stereo or only really 2.O ?
Is it 8GB of ROM or 6GB with a software cheat?
Are devices based on AMLogic designss really electrically safe, or have AMLogic and suppliers cut corners there too!
They have conspired to deceive and defraud.
And dismissal of their fraud is wrong..
(2016-09-01, 14:24)Theguyuk Wrote: [ -> ]Hello
I cannot agree with your dismissal of AMLogic cheating or comparing Soc GHz, and low price does not make fraud legal.
World commerce and trade depends on binding agreed terms.
A Intel and AMD CPU are not the same but everyone expects their CPUs to run at the claimed 2GHz. They expect 2GB of ram to be 2GB, not 1.5 GB.
A Ford, A Vauxhall, A BMW car are not the same but if their Speedometer says you are traveling at 100 miles a hour, you would not accept in truth the car to be doing 50 miles a hour, only.
Fifty dollars is fifity dollars and not 35 dollars and two pebbles.
AMLogic set out to cheat GHz and subroutines to cheat benchmark tests.
Are their Soc 5.1 stereo or only really 2.O ?
Is it 8GB of ROM or 6GB with a software cheat?
Are devices based on AMLogic designss really electrically safe, or have AMLogic and suppliers cut corners there too!
They have conspired to deceive and defraud.
And dismissal of their fraud is wrong..
Someone needs a tin foil hat
And a Full Body Condom just to be sure, to accompany the Tin Foil Hat.
Let me ask how much Diesel emissions testing, did all those thousand of premium VW dealerships do worldwide when selling cars to customers?
Nada, none. There is a degree of trust from the manufacturer with every product sold to a customer, no matter what it is.
Let put this into perspective shall we, these ARM based AML SoC's cost next to nothing compared to full monty Intel i series products. Which still cannot give you 10bit Hardware HEVC and HDMI 2.0 at a relatively affordable price.
WeTek did actually test the original Revision B. S905 a hell of a lot and found video decoding produced artifacts for 10bit 2160p HEVC Sat. TV streams. This was directly relevant for the S905 line of products they intended to sell and actually delayed the launch of the S905 equipped WeTek Hub. The CPU speed pretty much pales into insignificance, when all you need is a decent GPU/VPU that Hardware decodes nearly everything.