Intel's Core M Performance Is Erratic Between Devices 85
An anonymous reader writes: AnandTech noticed some odd performance disparities with Intel's Core M CPU, a chip designed to bring high-powered processing to thin, fan-less devices. After investigating, they found that how OEMs build their laptops and tablets has a far greater effect on Core M performance than it does for other chips. "When an OEM designs a device for Core M, or any SoC for that matter, they have to consider construction and industrial design as well as overriding performance. ... This, broadly speaking, gives the OEM control over several components that are out of the hands of the processor designers. Screen size, thickness, industrial design, and skin temperature all have their limits, and adjusting those knobs opens the door to slower or faster Core M units, depending on what the company decides to target.
In the Core M units that we have tested at AnandTech so far this year, we have seen a variety of implementations with and without fans and in a variety of form factors. But the critical point of all of this comes down to how the OEM defines the SoC/skin temperature limitations of the device, and this ends up being why the low-end Core M-5Y10 can beat the high-end Core M-5Y71, and is a poignant part of our tests. Simply put, if the system with 5Y10 has a higher SoC/skin temperature, it can stay in its turbo mode for longer and can end up outperforming a 5Y71, leading to some of the unusual results we've seen so far."
In the Core M units that we have tested at AnandTech so far this year, we have seen a variety of implementations with and without fans and in a variety of form factors. But the critical point of all of this comes down to how the OEM defines the SoC/skin temperature limitations of the device, and this ends up being why the low-end Core M-5Y10 can beat the high-end Core M-5Y71, and is a poignant part of our tests. Simply put, if the system with 5Y10 has a higher SoC/skin temperature, it can stay in its turbo mode for longer and can end up outperforming a 5Y71, leading to some of the unusual results we've seen so far."
Time to stop considering individual components. (Score:5, Insightful)
It's something some of us Apple fans have long figured out is that individual specs sometimes are completely meaningless.
Having a Core i7 will not actually feel more responsive in everyday tasks compared to a Core M if the i7 is paired with a spinning rust disk and the Core M has a PCI E SSD.
Similarly, just looking at the chip in the machine might not tell us everything if we don't know anything about how it's handling cooling or what specific design choices were made.
We're on the verge of reaching the 150HP car of computing. Don't really need much more for most tasks unless you're doing heavy lifting or looking to have fun, and even a lot of good clean fun can be had at 150HP.
Re: (Score:1)
Pretty much any serious computer enthusiast knows this. Generally only tech illiterates fall for this when they buy the prebuilt computers at Best Buy and the like.
Re: (Score:1)
That or sales pitches as well, Ive noticed that most manufacturers wont give mention to what your putting their parts on. A top of the line graphics card wont do you much good if your motherboards chipset is past its prime. All the parts work together, no matter how you slice it, something is going to have to be the weakest point.
Re: (Score:2)
It will do fine in your example, as the graphics card is directly connected to the CPU's PCI express interface.
The PC market is suprisingly free of rip offs these days, at least on the desktop side - and if you don't choose the low power CPUs while wanting something faster.
Even there you don't suffer horrible bottlenecks like in the past (PIO mode hard drive, not enough RAM, omitted L2, then Intel graphics using up limited FSB and memory bandwith, then the first gens of Celeron Pentium 4 - all other Celeron
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
So... Anything involving opening software, reading a file, saving a file, or any operation which hits disk cache.
IOW, he's correct.
Re:Time to stop considering individual components. (Score:4, Insightful)
If you're hitting disk cache that often, you need more memory.
Re: (Score:2)
That's why I compared it to modern cards which generally range from about 130 to 200 HP, with most ranging near 150 HP.
Simply put, we've got "sufficient" memory bandwidth and latency times that for most use cases, it doesn't matter.
My point was that we need to start evaluating machines based on the whole gestalt of the build and not just, "oh this has X, Y, and Z parts therefore it will perform in some certain way."
Re: (Score:2)
I don't think they do actually. When techies are screaming about how Apple products are overpriced, then start suggesting things that should have higher specs but the usability on them is shit because they built the device based on a checklist and not actually thinking about how anyone's going to use it...
Re: (Score:2)
Re:Time to stop considering individual components. (Score:4, Insightful)
How often do you edit multiple document without closing the word-processor in-between or loading up other application?
Because the user that logs in, runs Word, Excel, etc. and then doesn't close any of them until they shut down is a rare beast.
And let's not even get into the swap usage of doing something like that.
Disk performance affects everything you do on a modern machine, which is why SSD's are such a boon to any desktop. Hell, even things like event logs etc. are CONSTANTLY writing to disk in the background, even if the writes are cached.
And I think you'll find that the first thing that a lot of modern word-processors do is make a temporary disk copy of your document when you first open it, so you can edit without disturbing the original. That's how it's able to "recover" your unsaved work.
Disk access is a critical part. Not every single application will need it 100% of the time, but when disk access hits as the bottleneck, you will know about it.
I'm seriously considering scrapping planned RAM/CPU upgrades at my workplace this year and just dropping in cheap SSD's as they'll make TWICE the difference that even a bit more RAM would to the average desktop user's experience.
Re: (Score:1, Troll)
I am one of those users who thinks having to open and close an application every time I need it is idiotic.
At power up, everything stays open for days or weeks.
Do you want to know the best way possible to make your machine fast and long lived so that your disk performance isn't your bottleneck?
Put a crap ton of memory on the damned thing. Buy an older CPU, but stog it full of as much memory as you can afford, to the point that it seems like a ridiculous amount of memory.
Far too many machines suffer from ha
Re: (Score:3)
They did, but honestly, a memory upgrade can be the most bang for the buck and last in the long run.
If you take a machine which has used all of its memory, and is already paging ... everything else it does is going to be slow. The machine is now constrained by memory and IO speed, and always will be.
When my wife upgraded her work laptop to more memory, she was suddenly shocked she could launch task manager while the machine was
Re: (Score:3)
Isn't that what they call it in the olden days? Memory is a poor man's upgrade?
No, in the olden says memory was expensive. A populated 4 or 8MB (yes, MB) RAM expansion for an Amiga 500 or even 2000 could easily cost more than the computer. And the cost of three 8MB VME sun3/4 RAM boards for my 4/260 was the same as the cost of the chassis, mainboard, SCSI controller, disk, and tape... put together.
Which olden days were you thinking about?
RAM is now practically free, but practically no machines come with less than 2GB of it, which is fine for most people and most purposes unless you ha
Re:Time to stop considering individual components. (Score:4, Insightful)
LOL ... Troll??? Really? What morons are getting mod points these days?
Pointing out that "more RAM == faster computer" is not trolling.
Pointing out that some fucking idiot who can't count his own toes has mod points but should be drowned in his own drool? Now, that might be trolling.
Re: (Score:3)
You should do some testing ... because I'd be willing to bet the average user is going to see FAR more improvements from more RAM than faster SSD. Don't make paging faster, eliminate it.
Most users spend most of their time in the web browser, and the rest of the time either booting or hibernating. If you add RAM, hibernation takes longer. If you replace HDD with SSD, you improve the speed of everything. Since most users run Windows and Windows users need antivirus and that craps all over disk access times, disk access times are mortally important.
Most computers have 2GB or more RAM now, so most people can run one or two programs at once without swapping. That's broadly enough RAM for most p
Re: (Score:2)
Honestly, I find modern web browsers to be complete memory hogs.
I guess I just use machines differently ... I have six virtual desktops, 3 completely different browsers being used for different things (with multiple windows of multiple tabs each), iTunes, a couple of VMs. Basically I keep as many things open as I need.
As I said, my wife's work laptop was a dog slow machine, and by the time she booted, launched her firewall, opened Outlook and maybe one other thing .. the machine was already thrashing and s
Re: (Score:2)
Depends on the OS, actually.
Windows is Application-oriented, so you work on a document in an application, then when you're done with the document, you close the application more so than closing the document.
MacOS (and OS X) however are document-oriented - you work on
Re: (Score:2)
Depends on a lot of things but one thing for certain... swap on a SSD greatly improves system responsiveness when you have a lot of open applications (on any OS: Windows, Mac, Linux, BSD). Being able to page out anonymous memory to fast swap is a big deal. Nominal file storage on a SSD greatly improves program startup, boot times, photo and document handling. I've found though that it's really having swap space on the SSD that makes the biggest difference. I have a multitude of machines ranging from 1G
Re: (Score:2)
How often do you edit multiple document without closing the word-processor in-between or loading up other application?
It's not unusual for me to have InDesign, Acrobat, Photoshop, Firefox, Chrome, Picasa, and maybe some more programs open at the same time, and be switching actively between most of them. That's what it looks like when someone is getting work done on a computer.
And let's not even get into the swap usage of doing something like that.
Swap? Are you new?
Re: (Score:2)
The 32 Gig of memory in my desktop looks down upon swap and shrugs.
Re: (Score:1, Insightful)
Re: It is about balance and sufficient resources.. (Score:2)
Performance
Re:Time to stop considering individual components. (Score:4, Interesting)
Having a Core i7 will not actually feel more responsive in everyday tasks compared to a Core M if the i7 is paired with a spinning rust disk and the Core M has a PCI E SSD.
Cool - I'll transcode a 900MB .dv clip to h.264 with ffmpeg on my 4-core hyperthreadding i7 (the low-power model, even) with a simple drive mirror, and you run it on your Core M with a PCIe SSD (on a Mac even), and let's see when each job finishes.
(as usual, use the right tool for the task)
Re: (Score:3)
Now those two groups are moving apart content creators (programmers, video/audio editors, web site creators etc) will have t
Because Apple's tradeoffs are perfect, duh (Score:2)
So... why exactly do you need a PCIe SSD for watching videos again?
Because any above-average part that Apple includes is what makes their devices superior to the rest of the market, and anything they exclude or go for below-average on is superfluous, of course.
Re: (Score:2)
Because when everything pages out or in everything will feel smoother than using a sata SSD. If you're going to spend PCIE lanes on having the sata controller, might as well skip the controller and just use a PCI E SSD.
Re: (Score:2)
"It's something some of us Apple fans have long figured out is that individual specs sometimes are completely meaningless."
No they are not.
If you are transcoding video then a fast video card if using OpenCL or Cuda will really speed things up.
If you are not then an I7 will really help.
I love OS/X and I really like my macbook pro but the new iMac, Mac Book Pros, and Mac Pro are not as flexible as I want or need.
SSDs are advancing too fast for me to not want the option of installing a new one. Apple charges t
Re: (Score:2)
I said sometimes and also referenced everyday tasks. Are you transcoding video on a daily basis?
If you're a person who's transcoding video, then sure, one core i7 with less cores but higher frequency will do the job better than another core i7 with more cores and lower frequency.
But if you're looking for a machine to do office suite docs, browse the web, email, etc. then comparing machines based on does this machine have a Core M vs Core M becomes irrelevant. You have to think about things like build qual
Nothng new (Score:4, Interesting)
This has been the case for Intel CPUs for many years. Back in the Core 2 days they were already letting laptop manufacturers customize the power profiles (and therefore performance) of their mobile CPUs to suit the thermal load handling ability of their machines.
If you have one of those old Core 2 machines and don't install the Intel chipset driver for it (or install the generic one) it will get hot and loud pretty quickly. The only real difference now is that the CPU has better management built in and works okay without the driver giving it hints.
Better cooling = better performance (Score:3)
I can't see any problem with that.
Just read a few reviews before buying to make sure you get a device that is properly designed.
Re: (Score:3)
The problem is that the devices reviewed aren't the devices that you have available in the shops. Specifications and designs changes so fast that you can't keep up.
Re: (Score:2)
Truman (Score:1)
Fantastic... (Score:5, Insightful)
As though that isn't enough fun, now even expert level knowledge of the model numbers won't tell you how fast it is because the OEM can gimp it to suit their chassis design. It's a good thing that basically all modern CPUs are really fast, or this would be downright depressing.
Re: (Score:3)
Just bring a ruler, and take the thinnest one. Or the shiniest, depending on your preferences.
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
Lifting a *PDP*?! (Score:2)
Man, I got to use a PDP11 at my father's institute. I can assure you that lifting the machine would require industrial machinery. Also, the machine was rack-mounted with its disk... So that advice would not be very welcome :)
Re: (Score:2)
It was probably about the size of 4U rack (though it was not in a rack), the bootstrap I believe was input by a series of switches..... (it was not long - maybe 16 or so words) - I think there were 16 switches.... It would have been around 1978 (I believe I was 14 while doing some free work in the datacentre - though the big computers were around 100 miles away).... and it was probably not leading edge
Re: (Score:2)
No it was the Apple III. It had a few issues because the design was really pushing board design.
PDP?
DEC Would never allow that. Frankly picking up and dropping most PDP 11s could cause death or a minor earthquake.
Re: (Score:3)
[...] early Apple III computers where heat would cause chips to expand out of their sockets, [...]
“It’s not wise to upset an Apple III.”
“But sir...no one worries about upsetting a Droid.”
“That’s ’cause a Droid don’t cause people’s chips to expand out of their sockets. Apple IIIs have been known to do that.”
“I suggest a new strategy, Artoo. Let the Apple III win.”
Re: (Score:2)
I had a Commodore PET that did that.
Solution, lift the lid (it pivoted) and push down on all the chips. Lower the lid and power it up.
Re: (Score:2)
Amigas with DIP 68ks were a bit fiddly as well, once you'd de-socketed the CPU once or twice. And especially Amigas with accelerators which were plugged into their DIP 68k socket...
New Macbook (Score:2)
Re: (Score:2)
Re: (Score:2)
There are a number of things that could conceivably alter the results between tests of similar machines, such as ambient temperature, the specific video codec used for movie playback (some are more CPU intensive than others), whether the video was being streamed from the HD or streaming wirelessly (wireless radios suck up a lot of power), the screen brightness, and so on. I'd simply look at the range of numbers as best and worst case scenarios that you're likely to see.
I can't answer any of your other ques
Device design affects Intel's CoreM performance (Score:4, Insightful)
Fashion accessory (Score:5, Insightful)
Better cooling of CPU gives better results (Score:5, Insightful)
My experience with ULV, i7 ULV, Core M's (Score:1)
Wow (Score:1)
Re: (Score:1)
temperature difference / thermal resistance = power dissipation
5K/W heatsink at 20K above ambient dissipates 4W. Same 5K/W at 40K above ambient dissipates 8W.
False advertising, IMHO (Score:1)
If a buy a computer with a CPU that is rated at X GHz then that CPU had better be able to maintain that frequency, always. Otherwise it's a meaningless number. CPUs can already overclock themselves (Turboboost) above that frequency so if they can also legitimately underclock themselves then the 'rated frequency' is completely meaningless. I don't think that is acceptable. I encourage all slashdot readers to test their new computers under load and if they cannot maintain their rated frequency RETURN THEM! Or