Puma (microarchitecture)

Puma - Family 16h (2nd-gen)
Produced	From mid-2014 to present
Common manufacturer(s)	AMD
Max. CPU clock rate	1.35 GHz to 2.5 GHz
Min. feature size	28 nm
Instruction set	AMD64 (x86-64)
Cores	2–4
L1 cache	64 KB per core^[1]
L2 cache	1 MB to 2 MB shared
Socket(s)	Socket AM1 Socket FT3b (BGA-769)
Predecessor	Jaguar - Family 16h
GPU	Radeon Rx: 128 cores, 300-800 Mhz
Core name(s)	Beema Mullins
Brand name(s)	AMD APU

This article is about the microarchitecture. For the 2008 mobile platform, see AMD mobile platform#Puma platform (2008).

The Puma Family 16h is a low-power SoC microarchitecture by AMD. It succeeds Jaguar as a second-generation version, targets the same market, and belongs to the same AMD architecture Family 16h. The Beema line of processors are aimed at low-power notebooks, and Mullins are targeting the tablet sector.

Design

The Puma cores use the same microarchitecture as Jaguar, and inherits the design:

Out-of-order execution and Speculative execution, up to 4 CPU cores
Puma does not feature clustered multi-thread (CMT), meaning that there are no "modules"
Two-way integer execution
Two-way 128-bit wide floating-point and packed integer execution
Integer hardware divider
Puma does not feature Heterogeneous System Architecture or zero-copy^[2]
32 KiB instruction + 32 KiB data L1 cache per core
1–2 MiB unified L2 cache shared by two or four cores
Integrated single channel memory controller supporting 64bit DDR3L
3.1 mm² area per core
as a SoC (not just an APU) it integrates a Fusion controller hub with the following components: CPU, GPU, memory, and sometimes radios like wifi and cellular.

Instruction set support

Like Jaguar, the Puma core has support for the following instruction sets and instructions: MMX, SSE, SSE2, SSE3, SSSE3, SSE4a, SSE4.1, SSE4.2, AVX, F16C, CLMUL, AES, BMI1, MOVBE (Move Big-Endian instruction), XSAVE/XSAVEOPT, ABM (POPCNT/LZCNT), and AMD-V.^[1]

Features and ASICs

Features of AMD Accelerated Processing Units
Brand	Llano	Trinity	Richland	Kaveri	Carrizo	?	Desna, Ontario, Zacate	Kabini, Temash	Beema, Mullins	Carrizo-L
Platform	Desktop, Mobile			Desktop, Mobile	Mobile	Desktop	Ultra-mobile
Released	Aug 2011	Oct 2012	Jun 2013	Jan 2014	Jun 2015	2016?	Jan 2011	May 2013	Q2 2014	May 2015
Fab (nm)	GlobalFoundries 32 nm SOI			28		14/16nm	TSMC 40 nm	28
Die size (mm²)	228	246		245	244.62	?	75 (+ 28 FCH)	~107		TBA
Socket	FM1, FS1	FM2, FS1+, FP2		FM2+, FP3	FP4	AM4	FT1	AM1, FT3	FT3b	FP4
CPU architecture	AMD 10h	Piledriver		Steamroller	Excavator	Zen	Bobcat	Jaguar	Puma	Puma+^[3]
Memory support	DDR3-1866 DDR3-1600 DDR3-1333		DDR3-2133 DDR3-1866 DDR3-1600 DDR3-1333			?	DDR3L-1333 DDR3L-1066	DDR3L-1866 DDR3L-1600 DDR3L-1333 DDR3L-1066		DDR3L-1866 DDR3L-1600 DDR3L-1333
3D engine¹	TeraScale 2 (VLIW5)	TeraScale 3 (VLIW4)		GCN 1.1 (Mantle, HSA)	GCN 1.2	?	TeraScale 2 (VLIW5)	GCN1.1		GCN
	400:20:8	up to 384:24:6		up to 512:32:8		?	80:8:4	128:8:4		TBA
	IOMMUv1			IOMMUv2			IOMMUv1	IOMMUv1^[4]		TBA
Unified Video Decoder	UVD 3			UVD 4.2	UVD 6	?	UVD 3	UVD 4	UVD 4.2	TBA
Video Coding Engine	N/A	VCE 1.0		VCE 2.0	VCE 3.0	?	N/A	VCE 2.0		TBA
Power saving GPU	PowerPlay	PowerTune					?
Max. № of displays²	2–3	2–4		2–4	3	?	2	2		TBA
TrueAudio	N/A			✔^[5]			N/A^[4]			TBA
FreeSync	N/A			✔			?
Direct Rendering Manager / Mesa 3D driver^[6]^[7]	✔^[7]					WIP	✔

¹ Unified shaders : Texture mapping units : Render output units
² To feed more than two displays, the additional panels must have native DisplayPort support.^[8] Alternatively active DisplayPort-to-DVI/HDMI/VGA adapters can be employed

Improvements over Jaguar

19% CPU core leakage reduction at 1.2V^[9]
38% GPU leakage reduction
500 mW reduction in memory controller power
200 mW reduction in display interface power
Chassis temperature aware turbo boost^[10]
Selective boosting according to application needs (intelligent boost)
Support for ARM TrustZone via integrated Cortex-A5 processor
Support for DDR3L-1866 memory^[11]

Processors

Desktop/Mobile (Beema)

Family	Model	Socket	CPU				GPU			TDP	Memory
Family	Model	Socket	Cores	Frequency	Max. Turbo	L2 Cache	Model	Config.	Max. Freq.	TDP	Memory
A8	6410	Socket FT3b	4	2.00 GHz	2.4 GHz	2 MB	Radeon R5	128:?:?	800 MHz	15 W	DDR3L-1866
A6	6310			1.80 GHz	2.4 GHz		Radeon R4		800 MHz	15 W	DDR3L-1866
A4	6250J			2.00 GHz	N/A		Radeon R3		600 MHz	25 W	DDR3L-1600
A4	6210			1.80 GHz			Radeon R3		600 MHz	15 W
E2	6110			1.50 GHz			Radeon R2		500 MHz	15 W
E1	6010		2	1.35 GHz		1 MB	Radeon R2		350 MHz	10 W	DDR3L-1333

Tablet (Mullins)

Family	Model	CPU				GPU			Power		Memory
Family	Model	Cores	Frequency	Max. Turbo	L2 Cache	Model	Config.	Max. Freq.	TDP	SDP	Memory
A10 Micro	6700T	4	1.2 GHz	2.2 GHz	2 MB	Radeon R6	128:?:?	500 MHz	4.5 W	2.8 W	DDR3L-1333
A6 Micro	6500T		1.2 GHz	1.8 GHz		Radeon R4		401 MHz
A4 Micro	6400T		1.0 GHz	1.6 GHz		Radeon R3		350 MHz
E1 Micro	6200T	2	1.0 GHz	1.4 GHz	1 MB	Radeon R2		300 MHz	3.95 W		DDR3L-1066

References

1 2 "Software Optimization Guide for Family 16h Processors". AMD. Retrieved August 3, 2013.
↑ "AMD launches new Beema, Mullins SoCs". ExtremeTech. 2014-04-29. Retrieved 2014-05-02.
↑ "AMD Mobile “Carrizo” Family of APUs Designed to Deliver Significant Leap in Performance, Energy Efficiency in 2015" (Press release). 2014-11-20. Retrieved 2015-02-16.
1 2 Thomas De Maesschalck (2013-11-14). "AMD teases Mullins and Beema tablet/convertibles APU". Retrieved 2015-02-24.
↑ "A technical look at AMD’s Kaveri architecture". Semi Accurate. Retrieved 6 July 2014.
↑ Airlie, David (2009-11-26). "DisplayPort supported by KMS driver mainlined into Linux kernel 2.6.33". Retrieved 2014-07-02.
1 2 "Radeon feature matrix". freedesktop.org. Retrieved 2016-01-10.
↑ "How do I connect three or More Monitors to an AMD Radeon™ HD 5000, HD 6000, and HD 7000 Series Graphics Card?". AMD. Retrieved 2014-12-08.
↑ Shimpi, Anand. "AMD Beema/Mullins Architecture & Performance Preview". AnandTech. Retrieved 29 April 2014.
↑ Shimpi, Anand. "New Turbo Boost, The Lineup and Trustzone". AnandTech. Retrieved 29 April 2014.
↑ Woligroski, Don. "Meet The Mullins And Beema Tablet APUs". Toms Hardware. Retrieved 29 April 2014.

External links

AMD processors

List of AMD microprocessors

Discontinued

pre-x86	Am9080 Am2900 list Am29000 Berkeley RISC Alchemy MIPS32

Early x86 (16-bit)	Am286

IA-32 (32-bit)	Am386 Am486 Am5x86 K5 K6 K6-2 K6-III Duron Athlon Athlon XP Athlon MP

x86-64 (64-bit)	Sempron Athlon 64 Athlon 64 X2 Phenom Turion

Current products

IA-32 (32-bit)	Geode

x86-64 (64-bit)	Phenom II Athlon II Opteron AMD APU FX

Lists

Microarchitectures

AMD sockets and chipsets

AMD sockets

Desktop sockets	Super Socket 7 Slot A Socket A Socket 754 Socket 939 Socket AM2 Socket AM2+ Socket AM3 Socket AM3+ Socket FM1 Socket FM2 Socket FM2+ Socket AM1 Socket AM4

Mobile sockets	Socket A Socket 563 Socket 754 Socket S1 Socket FT1 (BGA413) 2011 Socket FP2 (BGA827) 2012 Socket FS1 Socket FT3 (BGA769) 2014 Socket FP3 (µBGA___) June 2014

Server sockets	Socket A Socket 940 Socket F Socket F+ Socket G3 (cancelled) Socket G34 Socket C32

Motherboard chipsets List of AMD chipsets

ATI chipsets	IGP 300 series/RX380 Radeon Xpress 200 CrossFire Xpress 3200 Radeon Xpress 1250

AMD chipsets	480X/570X/580X 690 Series 7-Series 8-Series 9-Series

Single-board computer & Single-board microcontroller

Devices

SoCs

ARM	Allwinner Ax Exynos i.MX K3V3 MT Nomadik NovaThor OMAP Rockchip Qualcomm Snapdragon Tegra WonderMedia

MIPS	Jz

x86/x86-64	Atom Jaguar-based Puma-based Quark

Software

Comparison of single-board computers

This article is issued from Wikipedia - version of the Tuesday, February 09, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.