Bay Trail benchmark appears online, crushes fastest Snapdragon ARM SoC
- Thread starter mikk
- Start date
StrangerGuy
Diamond Member
- May 9, 2004
- 8,443
- 124
- 106
OMG ARM is failing!!!!
How would consumers tolerate their ARM crap right now when this new chip would load their Facebook imperceptibly faster?
How would consumers tolerate their ARM crap right now when this new chip would load their Facebook imperceptibly faster?
SOFTengCOMPelec
Platinum Member
- May 9, 2013
- 2,417
- 75
- 91
If true, very ouch for INTEL indeed.
Surely, they should be comparing their unreleased stuff with the upcoming, new, considerably faster generation of Arms which are coming, such as the Cortex A57 like this. Unless I am missing something, here.
Which is 2 or 3 times faster (depending on what you count as the previous Arm series, and which benchmarks you use, anyway, faster than the previous generation Arms).
Also, the lower end of the market place, tends to be interested in other factors, rather than pure speed.
E.g. If your e-book reader (which already, almost instantly changes pages on request, and the tiny delay, is all because of how long the screen physically responds, i.e. NOT cpu limited).
Would you go out and buy a new one, because the new one has a 5% faster cpu ?
So, OUCH for INTEL, as they seem to be comparing the wrong generation of cpus, and other factors, such as price, battery consumption etc, may well be more important than outright speed, to the majority of the applicable market.
Even desktops these days, bought by non-computer enthusiasts and non-gamers, probably only need to be FAST ENOUGH, rather than 5% faster than everything else.
Surely, they should be comparing their unreleased stuff with the upcoming, new, considerably faster generation of Arms which are coming, such as the Cortex A57 like this. Unless I am missing something, here.
Which is 2 or 3 times faster (depending on what you count as the previous Arm series, and which benchmarks you use, anyway, faster than the previous generation Arms).
Also, the lower end of the market place, tends to be interested in other factors, rather than pure speed.
E.g. If your e-book reader (which already, almost instantly changes pages on request, and the tiny delay, is all because of how long the screen physically responds, i.e. NOT cpu limited).
Would you go out and buy a new one, because the new one has a 5% faster cpu ?
So, OUCH for INTEL, as they seem to be comparing the wrong generation of cpus, and other factors, such as price, battery consumption etc, may well be more important than outright speed, to the majority of the applicable market.
Even desktops these days, bought by non-computer enthusiasts and non-gamers, probably only need to be FAST ENOUGH, rather than 5% faster than everything else.
Last edited:
Snapdragon 800 is the wrong generation? Can I buy Snapdragon 800 tablets right now? Intel didn't compare, it's just a leaked (low clocked) development Bay Trail-T system from someone. It could be any OEM.
You are wrong Ancalagon44...
CloverTrail+ Atom scores at 1ghz 0.10 points in Cinebench 11.5 single thread
Jaguar(Temash/Kabini) scores at 1ghz 0.26 points in Cinebench 11.5 single thread
Bay Trail T is 50% faster than CloverTrail+ thus it should score at 1ghz 0.15 or 0.16 points in cinebench 11.5 single thread
Top of the line Bay Trail GPU will be as powerful as the lowest end Temash A4-1200 GPU, Bay Trail will have an egde over Temash/Kabini in power consumption while this could be gone with "Beema" that arrives in 2014.
as for Bay Trail T scores, does the benchmarking program detects when the chip is using boost/turbo and show it or does it not?
Since Bay Trail T chip could have turbo/boost mode.
edit:
Lets not also forget how easily benchmarks can be faked, we are talking about antutu benchmark and there were tons of cases of fake benchmarks so please take this Bay Trail benchmark with a grain of salt!!!
Last edited:
SOFTengCOMPelec
Platinum Member
- May 9, 2013
- 2,417
- 75
- 91
The ebook reader is a VERY bad example for battery consumption, as the e-ink type, usually consume zero power (relatively), between page flips, hence potentially offering a battery life, measurable in months, or certainly long periods of time.
The upcoming Arm chips, also offer phenomenal battery consumption improvements, (relative to performance).
Anyway, I need to back out of significant disagreements between Intel and Arm stuff, because I am not very well read up on what is happening, with them.
I'm partly skeptical of early benchmarks, these days.
I prefer to wait and see what gets released, how much it costs, how fast it is, and then make my own mind up. It is too speculative at this time. (E.g. Haswell was a while ago, optimistically expected in some quarters, to be something like up to 20% IPC faster, and have soldered TIM).
Sometimes, the better/faster/favourite cpu is way over priced, especially if Intel is the seller, compared to the still fast, but not best alternatives.
Last edited:
podspi
Golden Member
- Jan 11, 2011
- 1,982
- 102
- 106
Isn't A57 expected to have higher power consumption than even A15? It looks to me like Intel has a winner. Can't wait for Windows tablets with these to be released, looks like a great upgrade from my Clovertrail one 
BTW, I am sure the CPU was using boost and these aren't the results @ 1.1ghz. That would be amazing, but I doubt it, given what Intel has already said about BT.
BTW, I am sure the CPU was using boost and these aren't the results @ 1.1ghz. That would be amazing, but I doubt it, given what Intel has already said about BT.
As Intel has been stating for a long time, it'll be 50-60% faster than Clover Trail. Intel does not make node improvements on this level, so it isn't any faster than it was last slide we saw which clearly states 50-60% faster.
It'll be a nice chip but nobody at Qualcomm or AMD is overly worried, particularly as the graphics performance will still only be half as fast.
It'll be a nice chip but nobody at Qualcomm or AMD is overly worried, particularly as the graphics performance will still only be half as fast.
SOFTengCOMPelec
Platinum Member
- May 9, 2013
- 2,417
- 75
- 91
Quote is from my link in my first post.
As with some other cpu solutions, you can choose between less performance, but very low power consumption, or huge performance, with a power consumption penalty, on the fly, depending on what the device is doing.
Therefore it can offer a huge battery life with the option to switch to "full" speed.
If Intel are so wonderful, how come a significant percentage of the existing/current small devices market is NOT Intel cpus based ?
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 145
- 106
You do know that some ARM designers, like Qualcomm and Apple rejects big.LITTLE design? It seems to rather be the solution for companies with limited R&D to pick big.LITTLE.
I predicted many months ago this would happen, once intel focused on the task at hand - they produced a part which destroys all ARM SOCs within a lower power envelope on their 22nm process. It's pretty hilarious considering that the ULV Haswell gets 13 hours of battery life in a macbook air while the iPad 4 gets what? 10 hours on a custom ARM SOC? So based on that, I wonder how much battery life bay trail will get? 15+? Probably.
Personally, I think it's freakin hilarious that ARM's hubris is about to catch up to them. They said intel would never ever beat them in efficiency. Hilarious. I can't wait to see the silly damage control posts full of people doubting intel.
Nothing will shut ARM up faster than seeing their best ARM SOCs get destroyed, easily, by intel's bay trail.
Personally, I think it's freakin hilarious that ARM's hubris is about to catch up to them. They said intel would never ever beat them in efficiency. Hilarious. I can't wait to see the silly damage control posts full of people doubting intel.
Nothing will shut ARM up faster than seeing their best ARM SOCs get destroyed, easily, by intel's bay trail.
Last edited:
SOFTengCOMPelec
Platinum Member
- May 9, 2013
- 2,417
- 75
- 91
Forget me.
Let's give a higher priority, to letting the main experts themselves know (Arm themselves).....
Here is their latest details about it, including that the A57 CAN have it, see other link for details
I.e. Why are they (Arm) going on and on about it, if it is a poor solution ?
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 145
- 106
So Qualcomm and Apple is...incompetent because ARM PR states otherwise?
Who is using the least amount of money on R&D of ARM chips. ARM Holding, Qualcomm or Apple?
Because the old Atom uArch was pretty crappy, is made on an older process than ARM competitors and also slower multi-threaded and especially the GPU sucks.
Baytrail is a complete new uArch and hence it will be much, much better. I've been saying it for weeks. Just look at the RAZR I vs RAZR M. These are identical phones excpet the SOC. I with single-core! Atom the M with MSM8960 (dual-core Krait at 1.7 Ghz).The crappy old Atom based phone already beat the ARM based one in terms of CPU performance and battery life. So it is no wonder that the new uArch on a process advantage leads to a serious advanatge for intel. It is just common sense...
A57 is not for phones. It competes with haswell Y series in terms of TDP and not with atom. A15 is already pretty power hungry. Intels version of big.little is turbo and especially low power states and fast switching between them.
Last edited:
I'm under the impression that big.Little is completely worthless for consumer level products such as tablets and smartphones, while it *can* be useful in highly parallel server environments. This is why Qualcomm and Apple do not, and likely will never have, plans to use big.Little. That's my impression anyway, someone correct me if i'm mistaken.
Last edited:
SOFTengCOMPelec
Platinum Member
- May 9, 2013
- 2,417
- 75
- 91
I'm not trying to say (when I first referred to this), that it is the best, good, cheapest for R&D or anything else.
I was just trying to explain that Arm ALSO has got techniques for making LOW power consumption chips.
I have to admit, I CAN'T understand the big.little concept, myself either.
It seems terribly wasteful to have 8 cores, just so that you can have 4 high speed cores or (only 4 max on at same time) 4 low power, low speed cores.
Why can't they just have ONE core type, and either lower the clock frequency down, to reduce performance BUT significantly reduce the power consumption, and/or switch off stuff in the cpu (power gating), in order to conserve power. Much like Intel does with later parts, to reduce the power consumption.
But it does make a little bit of sense to me, because I have read that going from in-order, to out-of-order instruction execution, involves a considerable increase in power consumption, compared to the performance increase.
---------------------------------------------------------------------
Sadly (if you are an Intel Fanboy), or happily (if you are an Arm Fanboy), or annoyingly (if you are a girl, and my fanboy naming scheme is missing you out), there are much more complicated reasons, why Intel may have a major uphill struggle to get into the new markets.
The latest/greatest fab plants which Intel use, are also the MOST expensive. This would make it VERY difficult for Intel to offer the cheapest parts, while at the same time making them the best parts, and making the greatest profits.
But, maybe they can pull it off ?
Last edited:
Hans de Vries
Senior member
Watch out to not get fooled by the usual benchmark manipulations.
Marketeers won't lie outside the lawyer determined boundaries.
They'll let the journalists and naive fans do all the lying for them.
AnTuTu and Quadrant came into focus when Microsoft choose Nvidia's
Tegra3 for surface tablets as a direct competitor.
For instance this the Atom in the Motorola Razr-i scores better as the
Nvidia Tegra3 in Quadrant but if you go one step further and look at
the subscores in detail then you'll see that the Atom CPU score is just
1/3 of that of the Quad A9 Tegra3.....
source.
The 2GHz Atom with hyper-threading score is second in the list and the
CPU sub-score is the blue piece of the bar.
Furthermore. A benchmark as AnTuTu can easily be abused with
processors that can run in burst-mode at 5+ times or so their
(advertised) TDP. The benchmark is to short to warm up the phone
if you run at a far higher burst mode TDP.
Here's a simple example how the score degrades if you run AnTuTu
multiple times:
source.
And this is with an Arm device without all the high TDP super bursting
features of the Baytrail SOC...
Hans
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 145
- 106
When you got a 250M$ R&D budget for multiple core designs. This is what you get.
http://arm.com/files/pdf/Earnings_Tables_2012.pdf
Useless because it's a different benchmark. Also the Razr-i has a 1/2 Atom which is even much slower than Clovertrail+ 2/4, at least when all threads are used.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter