Intel ULP SoCs

ToTTenTranz said:
I don't get you guys. Is web browsing all you do with your smartphones or do you just happen to stop reading the first two paragraphs of each page in the review?
Click to expand...

It annoys me in a today's smartphone and yes I use the internet a LOT on mobile devices, most importantly to synchronize my calendar and its appointments/notes with all other mobile devices or update weather reports on a constant basis and that for my work primarily. Yes I read the entire thing and it still doesn't change the lifted eyebrow/criticism for the browsing battery life.

I said it has an oustanding price/performance ratio overall; what else do you want to read, that it managed to knock me out of my socks? It doesn't. I'd be actually willing to pay a tad more for a LTPS display if it would have had one.
 
Ailuros said:
It annoys me in a today's smartphone and yes I use the internet a LOT on mobile devices, most importantly to synchronize my calendar and its appointments/notes with all other mobile devices or update weather reports on a constant basis and that for my work primarily.
Click to expand...

And none of that is web browsing.
Web browsing power consumption varies immensely, as device makers usually have to make the decision to either make the browsing experience snappier and gobble away more power, or just limit the clocks on the CPU cores and get more battery time. It even varies a lot depending on the browser you're using, for the same device. It's also highly dependent on several optimizations like GPU acceleration.

I'm not expecting anyone to praise the Zenfone 2 like the second coming of the Messiah. It's a great smartphone for the price and that's it.
But the fact that its battery life on web browsing isn't very good, it doesn't mean its battery life is bad overall. On the contrary, battery life seems quite competitive with today's flagships with 5.5"+ screens like the Nexus 6, OnePlus One, Galaxy Note 4, etc.
 
ToTTenTranz said:
And none of that is web browsing.
Click to expand...

True; however there's no other measurement for other applications that use the internet either in the review nor do we know the source of the actual problem. Let's say the modem sucks and it consumes more power than it should; now what?

Web browsing power consumption varies immensely, as device makers usually have to make the decision to either make the browsing experience snappier and gobble away more power, or just limit the clocks on the CPU cores and get more battery time. It even varies a lot depending on the browser you're using, for the same device. It's also highly dependent on several optimizations like GPU acceleration.
Click to expand...

Why should I care as a potential consumer again? I consider at least 10 hours battery life with web browsing a must for today's =/>mainstream smartphones.
 
"Web browsing" is such a nebulous term, and 10 hours is such a long time to have to do any single activity on a phone between charges, that I'm not sure if you're serious or not.
 
Rys said:
"Web browsing" is such a nebulous term, and 10 hours is such a long time to have to do any single activity on a phone between charges, that I'm not sure if you're serious or not.
Click to expand...

Anyone would hardly browse for a constant 10 hours in a row on a smartphone. If however someone loses hypothetically N% of battery life for X online task and I lose N+30% for the same task, then my phone isn't obviously more efficient.
 
That's kind of my point I guess. While these "battery life" tests attempt to normalise the workload across the devices, and in some cases the person doing the analysis also tries to normalise the environment, by doing so, but also because they can't go far enough doing that (semiconductors for portable devices have active power management and wide ranges of dynamic power consumption), the end numbers are useless for you as the consumer.

So making a call on whether something has shit battery life or not can't be done from graphs. The best indicators of battery life for the consumer actually tend to come from the people that do a "day in the life" analysis (ideally over multiple days, in multiple locations).
 
True; however we've been chewing on the fact of how questionable a LOT of metrics are in the ULP mobile SoC world also forever and not just for browsing battery life. We all realize that what we see are just indications for each case, however those indications are rarely completely baseless in real time usage also.
 
Rys said:
"Web browsing" is such a nebulous term, and 10 hours is such a long time to have to do any single activity on a phone between charges, that I'm not sure if you're serious or not.
Click to expand...
I'd go even further and state that if you're browsing the web in a smartphone for longer than 2 hours every day, you obviously chose the wrong device for your daily activities.
 
ToTTenTranz said:
I'd go even further and state that if you're browsing the web in a smartphone for longer than 2 hours every day, you obviously chose the wrong device for your daily activities.
Click to expand...

But what else are you going to do on a smartphone for hours a day?
 
Talk, 9GAG, news feeds with dedicated app, alarm, calendar appointments, GPS navigation (I'd be a useless human being in the city without that), taking pictures of colleagues falling asleep to ask for ransoms, listening to music, watching short youtube videos from Emergency Awesome and playing chess.
Web-browsing is more of a 3/5-times-a-week-during-5 minutes-as-an-emergency kind of thing for me.
 
Albuquerque said:
Not sure who has seen it yet, but Intel has released details on a new Compute Stick that is using the Skylake ULP: http://arstechnica.com/gadgets/2015...-beefed-up-core-m-compute-stick-with-skylake/

Now we're talkin!
Click to expand...
We'll talk after we see reviews: the previous compute stick based on Z3735F seems to throttle a lot despite active cooling (see this review for instance). I somehow doubt that Skylake even on 14nm can consume less than a Z3735F unless its frequency is severely limited. But I'd like to be surprised :)
 
ToTTenTranz said:
I'd go even further and state that if you're browsing the web in a smartphone for longer than 2 hours every day, you obviously chose the wrong device for your daily activities.
Click to expand...
Not if you are commuting to work (train/metro to a big city center, taking 45 mins per direction + other casual use) and use Galaxy Note or iPhone 6+ for web browsing. Both are more than adequate for that task. I have an iPad Air at home, but don't like to carry a bag, so iPad is not an option (and I don't want to pay for secondary 4G data plan).
 
what's the power drain of the Z3590 and has anyone an estimate how that would compare to some top notch ARM cores? (maybe 801 or 805 or 810). Some years ago everybody said there is no way intel could run on a phone due to the power drain of the CPU alone and you had to get the LTE modem etc. all separate (thus it was tablet-ware in best case). I think it's now all integrated in Z3590, right?
Intel's http://ark.intel.com/products/91319/Intel-Atom-Processor-Z3590-2M-Cache-up-to-2_50-GHz gently skips the power line, not even the SDP.

I'd also wonder how the min power drain compares. Some Qualcomm guy once said, they can gate their snapdragons (32bit back then) to be as power efficient as a low power "little" core if there is not much going on. Now even Qualcomm runs big.little, I wonder if the Intel power states are in the area of a "little" core (I'd assume that most use cases have peak high performance demands, but most of the time, like browsing, it's rather that "idle" area that accounts for the average power drain at the end of the day).
 
You must log in or register to reply here.
Back
Top