AppleScruff1
macrumors G4
I'll give it a shot. 😀😀
Die Samsung Die!!
I hope that Samesung copycats go out of business.
I'll never buy anything Samsung makes. Except, of course, an Apple product that uses Samsung components.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Samsung to Produce A9 Chips for Apple's Next-Generation Devices
- Thread starter MacRumors
- Start date
- Sort by reaction score
I'll give it a shot. 😀😀
Die Samsung Die!!
I hope that Samesung copycats go out of business.
I'll never buy anything Samsung makes. Except, of course, an Apple product that uses Samsung components.
WAIT. WAIT! I really don't think these guys understand whats happening here. The new iPhone may simply be the beast of beasts. The A8X is insanely powerful, now were getting the A9 on a frekin 14nm Process, but more so there graphics processor should be a version of the PowerVR series 7xt, which is said to be as powerful as the Desktop Nvidia GTX 750 TI, which is also said to be more powerful than both the XBOX ONE, and the PS4. Insane, just insane.
If this processor, and graphics go into a new AppleTV, well really have an actual full console that is at least as powerful as both the XBOX ONE, and PS4. Agian, INSANE.
chrmjenkins
macrumors 603
Even in a 16 cluster configuration, a 7XT is only 614 GF (FP32) at 600 MHz. It also has vastly inferior memory bandwidth compared to consoles. It's going to take a lot more time to equal current gen consoles.
Theoretically, yes, but it seems Apple currently doesn't care about the Apple TV, sadly. They don't really understand the potential it has. They just don't get it.
We can at least hope that the A9 SoC finally has the following:
- Full hardware encode/decode for H.265 and 4K@60fps
- HDMI 2.0 and HDCP 2.2
NV already does all that with GM206 and IMGTec already has the stuff ready to integrate. Since there mm² needed for that would be minimal on 14nm, that is a no-brainer.
If Apple launches a new Apple TV, they should do their homework and also adapt their content to H.265 (4K). iTunes is seriously behind already, with Netflix and Amazon already offering that.
There's so much potential in 14nm, it's insane. This will be a very powerful node and it seems that Samsung has a clear winner here.
To what end? They make no use of this power aside from a few games. Basically doesn't each generation just end up compensating for the next IOS release?
Clearly I've been spoiled by my OnePlus One. Allow me to lower my standards.
Its a FAB. There just putting the chips together.
Its a FAB. There just putting the chips together. I doubt they will know exactly how Apple put together the design.
Remember both Samsung, and TSMC manufactured the A8. And the A7 was mostly Samsung, and there still struggling to make a true 64bit Processor.
Its a FAB. There just putting the chips together. I doubt they will know exactly how Apple put together the design.
Remember both Samsung, and TSMC manufactured the A8. And the A7 was mostly Samsung, and there still struggling to make a true 64bit Processor.
NO! Not at all. There are lots of new IOS apps that are insanely powerful, and need lots of power.
Pixelmator for ios is almost a full Photoshop replacement. And Replay is a frekin self editing video app, that needs allot of power to run.
But for the ordinary punter, they make no use of these speed bumps, no multi-tasking etc. How many phone users would even touch these Apps? I am genuinely interested as I have had each generation of ipad and each one seems as snappy as the last. I know older modes, would we slower with a newer OS, but each generation claims 2x this and 4x that, but the fact remains for 99% of the time there appears to be discernible difference
Clearly I've been spoiled by my OnePlus One. Allow me to lower my standards.
Right... That's why your phone has exactly 40min less web browsing than the 6+ in tests: 13h45m. So, just about a wash, much superiority there! (sic). The 6, with the much smaller battery browses for 11h30min.
You can kill the battery of ANY phone in less than 12h if you really want to.
Eventually people will have chips implanted.
Guaranteed some people would do this now already.
IJ Reilly
macrumors P6
Not in Intigrated Circuit Design. Apple's A Series are designed In House at their Orlando, FL design Center.
Samsung, and TSMC Fab the Integrated Circuit Designed by Apple based on ARM. Global foundries uses Samsung's 14nm Process overseen by Samsung. TSMC can't cut it at 14nm at this point.
Samsung simply can't copy the Apple design. It's not in the same realm as swipes, pinches, and zooms. Or, exterior design.
Just like Chevrolet could not simply copy the wonderful 8 speed ZF AT in the 2015 Z06 and call it a THM800R5.
Samsung is a HUGE Co. Samsung Mobile was always the object of SJ's Thermo Nuclear Temper. Not their Fabs.
The A9 will be "Magical". 😉
I am driving at something entirely different. Apple is currently stuffing huge sums of money into the coffers of a direct competitor. This is very far from being an ideal business practice. I can well imagine that many discussion around the big table at Apple are about how they can avoid sending billions to Samsung, which they in turn use to fuel their competition in Apple's most important product market. This factor is way more immediate than those glacially-paced lawsuits, which for competitive purposes, are all but meaningless.
Darrensk8
macrumors 6502
You know, I see posts like this quite often and I wonder what I'm doing wrong, I'm lucky to get much over 7 hours usage from my 6 plus.....of course if I don't use it I'll easily get 1.5 days but that kind of defeats the purpose 😕
Last edited:
Glideslope
macrumors G3
Yes, I get that. Apple has no choice. They are not going to get into Fabrication.
The billions to Samsung's Fab Division (They have over 50 Divisions)
is an operating cost. It's part of doing business. Not everything is "Rainbows and Unicorns." I'm sure Apple wishes there were other options at this time. There are not. There will be in a few more years as others step up their Fabs. Reality however is if Apple wants the most secure supply chain they still need Samsung in a big way. The lawsuits with Samsung are done. Tim's first job was to put out the "Thermonuclear" SJ War and go about it in small bites, and much more methodically. Tim is not SJ. Apple is now Tim's. It will never be the same as it once was under SJ, and Android will soon be the "Vista" of the Mobil OS World.
Samsung also needs Apple in a big way.
That is actually a good point. However, there are at least two reasons why it doesn't really matter for Samsung who uses Samsung's capacity for 14nm (until at least 2017, where then the new 15 billion fab is ready for production, too, which also needs to be utilized heavily).
First, if Apple won't use up the capacity, Qualcomm, NV, AMD etc. will start to use it earlier, because Samsung needs to make money on it or either they can't finance their equipment, which is extremely expensive. Meaning that everyone will eventually use the best process for their high-end offerings, which is Samsung's.
Second, and knowing Apple, they already knew the price for TSMC's 16FF+ offering beforehand (since they always need to consider both semis) and they can surely dump the prices as low as they can be for Samsung to still make it worthwile for both companies. You could also spin it backwards and say that for Samsung, Apple is probably the worst customer they can get, because even when they may bring a guaranteed volume requirement, they are well known for the sheer price dumping methods and that is never an optimal position to be in. Without knowing how much Apple is paying Samsung, this is all speculation, anyway.
Apple knows that without Samsung, they will only have access to the second best, and they need access to 14nm to stay at the cutting edge so that Schiller has something "amazing" to sell to the crowds this and probably up until next year.
Let's hope Apple in turn doesn't just jump to new nodes, but also innovates, like they did with Swift and Cyclone. A quad-core Cyclone would be nice, but on 14nm there's so much headroom to improve IPC and go quad at the same time and still have lots of room for other improvements (image DSP, audio/video signal processing, security related stuff...). Also, don't forget that the M9 could also make use of 14nm, which would make it even more worthwhile.
The Apple Watch also really could improve drastically (battery life, idle power, more stuff that saves power directly integrated into the Watch SoC instead of using the iPhone as a remote device over WiFi/bluetooth, which is a total energy waste) by utilizing 14nm, which will probably be their 2nd Gen announced a year later. This is one of the reasons why I will stay away from the 1st Gen Watch. This will only bring the Watch to the masses, but the real potential can only be used with 14nm and beyond.
Last edited:
dampfnudel
macrumors 603
Do you have an inside source because I find it hard to believe that Apple would give the iPad mini 4 the A8, which is not too much more powerful than the A7. Why do you think the iPad Air 2 got an enhanced version, the A8X? I think all iPhones/iPads released this year, with maybe the exception of a possible iPhone 6c, will get the A9. Maybe the A9 will even be powerful enough to power the iPad Air 3 and Apple won't have to enhance it like they did with the A8. Like someone else said earlier in the thread, the A9 will be a beast.
The A9 (like the A8 was) is reserved for the cutting edge smartphones and there's VERY limited production capacity for the A9 in 2015. Apple already started producing the A9 at Samsung, because they need absolutely everything Samsung can make for their iPhones.
It's not like there is a choice. You will always use your high-end SoCs where they make the most sense first. The A8X is a different SoC that is solely targetted for the bigger iPads, which also have a way larger battery then the iPhone ever will. The bigger iPads is where Apple can go all-out, but iPhone is where it actually matters for Apple. Their whole strategy is based on iPhone first.
The A8 at least technologically is good enough for the mini, don't get your hopes up. If Apple decides otherwise, then because they want to broaden the range of sold devices with new technology that only the A9/A9X will have. But this won't happen too soon, since Samsung can only produce a limited amount of SoCs this year, their wafer output is limited and so is the yields of 14nm in the first year in mass production.
Last edited:
s2mikey
Suspended
IJ Reilly
macrumors P6
Since you both made more or less the same argument... Apple clearly doesn't have a choice, but that's exactly the problem. They are in the unhappy position of being forced to fund a competitor, and I don't think there's much question about it not being to the tune of many billions. I'm sure this matters a lot to Apple, probably orders of magnitude more than those lawsuits.
Yes, better to be Apple's subcontractor (with all the implications) than having no money at all...
Renzatic
Suspended
1nm is probably the bare limit of silicon, which is where we're expected to be around 2030 or so. Anything smaller than that will require an entirely new technology.
There economic limitations at play here, not just technical, that may force a switch of tech before that.
Renzatic
Suspended
Yeah, if these CPU companies can find a way to get to a smaller die for cheaper, they'll take it. Right now, graphene is looking like the next big thing in computer tech. It's expensive to produce right now, though, and not as easily worked with. But if someone finds a way to mass produce it for the same cost as what we're using now, everyone will jump on board with it long before we reach the theoretical limits of silicon.