Thursday, January 24th 2013

AMD "Richland" Desktop APU Lineup Detailed

AMD's A-series "Trinity" line of APUs may have helped make the APU outsell CPUs in 2013, but it won't be long before they're replaced by the new "Richland" A-series APUs for desktops and mainstream notebooks. "Richland" is a tweaked version of "Trinity" which sees AMD stick to the 32 nm process, and retain the "Piledriver" CPU micro-architecture, but increase CPU clock speeds, add a faster DDR3-2133 MHz dual-channel IMC, and integrate a Radeon HD 8000 series Graphics CoreNext iGPU into the silicon. Desktop APU models will take up with A##-6000 series numbering scheme.

The series will be led by AMD A10-6800K, which features every component on the "Richland" silicon unlocked, which includes two "Piledriver" CPU modules amounting to four x86-64 cores, and all stream processors on the iGPU unlocked, with the highest CPU and iGPU clock speeds enabled in the lineup. The iGPU model for this chip is Radeon HD 8670D. The A10-6800K features unlocked multipliers, making overclocking a breeze. Trailing it is the A10-6700, which features all physical components unlocked, but with slightly lower clock speeds, and locked BClk multipler. It features the same iGPU as its bigger sibling, the HD 8670D.

The AMD A8-6000 series consists of the A8-6600K and A8-6500, both of which feature four CPU cores, but slightly toned down iGPU cores, labeled Radeon HD 8570D. The A8-6600K features unlocked BClk multiplier. The next APU in the lineup is the A6-6400K, which is dual-core, features a further scaled down iGPU, bearing the model number Radeon HD 8470D, and unlocked BClk multiplier. Lastly, there's the A4-6300, an entry-level dual-core APU with Radeon HD 8370D graphics. All models with -K extension feature rated TDP of 100W, others 65W.

Moving on to the platform itself, it's known that "Richland" APUs will be built in the same packages as "Trinity," and as such existing A55, A75, and A85X chipset-based motherboards should be able to run them with BIOS updates, yet AMD plans to launch a trio of new FCH chipsets. Leading the pack is the A88X (eight SATA 6 Gb/s ports), followed by A78 (six SATA 6 Gb/s ports), and A68 (probably four SATA 6 Gb/s ports, entry-level).
Source: Expreview
Add your own comment

64 Comments on AMD "Richland" Desktop APU Lineup Detailed

#26
cdawall
where the hell are my stars
SteevoThey showed these running a full 1080P tablet gaming without issues. So the thing is, its going to dominate the market when you don't need a second GPU to play games on in a netbook, or tablets that are actually capable of all the cool shit we have been sold on, and even thinner and smaller PCs.
I am waiting to update my netbook specifically for these APU's to drop. I have been watching the rumor mill for a while on them (since the A10-6800K came up almost 6-8 months ago). Finally something worthwhile to upgrade both the CPU and GPU version on my ancient netbook along with a freaking 1366x768 or higher LCD...:rolleyes:
Posted on Reply
#27
Eagleye
Trinity had the capability of PCIE-3.0 but was never validated for time/cost/performance reasons It was said.
Posted on Reply
#28
Cortex
IGP still slower than 7750, Hybrid CFX slower than 8750.

Impressive, AMD you have fastest IGP world has ever seen, too bad it's glued to slow power hungry Pile of <beep> processor.


:D
Posted on Reply
#29
cdawall
where the hell are my stars
CortexIGP still slower than 7750, Hybrid CFX slower than 8750.

Impressive, AMD you have fastest IGP world has ever seen, too bad it's glued to slow power hungry Pile of <beep> processor.


:D
Depends on your application. Do you use CAD? Metro2033? heavy 3D multi tasking? Because all of those applications it performs better than its competition. Don't believe the hype a lot of reviewers have admitted they were wrong recommending the 3570K over the 8350.
Posted on Reply
#30
jagd
I LOLed ,i have an offer for you ,you can be 2nd coming of Jesus .You only need to have intel's marketing budget :roll:
KonceptzBut when Intel does the same thing its the 2nd coming of Christ :confused:
Posted on Reply
#31
NeoXF
I'm surprised no one is commenting on the supposed new chipsets as well... And why does AMD feel a need for them... I for one hope they bring better clocking for the CPU and RAM, it's gonna be a pity if most people couldn't even clock their RAM at Richland's default clocks. On the same note, why no new chipsets for AM3+...

Seeing as how faster iterations of L3-less Vishera CPUs (no arch improvements at all?) are possible so soon, I'd imagine FX series CPUs might get a refresh soon too...
Posted on Reply
#32
Frick
Fishfaced Nincompoop
cdawallDepends on your application. Do you use CAD? Metro2033? heavy 3D multi tasking? Because all of those applications it performs better than its competition. Don't believe the hype a lot of reviewers have admitted they were wrong recommending the 3570K over the 8350.
Who have done that? Would be an interesting read.
Posted on Reply
#33
ur6beersaway
Was wondering what socket for richland (FM-3)? yawn
Posted on Reply
#34
Dent1
CortexIGP still slower than 7750, Hybrid CFX slower than 8750.

Impressive, AMD you have fastest IGP world has ever seen, too bad it's glued to slow power hungry Pile of <beep> processor.


:D
You mean AMD has the fastest APU the world has ever seen :)
NeoXFSeeing as how faster iterations of L3-less Vishera CPUs (no arch improvements at all?) are possible so soon, I'd imagine FX series CPUs might get a refresh soon too...
I would love to see a small Piledriver tweak before moving to Steamroller.
Posted on Reply
#35
newtekie1
Semi-Retired Folder
I'd like to see some of these make their way into the 13.3" Ultrabook form factor laptops. I'm annoyed that the only Ultrabook out there currently that is even the slight bit capable of gaming is $1,300 and has to have a dedicated GT620M along with the iGPU which kills battery life.
Posted on Reply
#36
NeoXF
ur6beersawayWas wondering what socket for richland (FM-3)? yawn
Ugh...

newtekie1I'd like to see some of these make their way into the 13.3" Ultrabook form factor laptops. I'm annoyed that the only Ultrabook out there currently that is even the slight bit capable of gaming is $1,300 and has to have a dedicated GT620M along with the iGPU which kills battery life.
AMD wrote in the fine-print of a press release that it's mobile quad-core 19W parts are capable of 1100 points in 3DMark11 Performance test. Models tested where (Trinity) A8-4555M with 780P points vs (Richland) A8-5545M with 1100P points, both 19W TDP.
Posted on Reply
#37
Aquinus
Resident Wat-man
newtekie1Ultrabook form factor laptops
Doesn't Intel own the trademark of "Ultrabook" so don't they have a certain level of control over what devices can be called Ultrabooks?
I know what you're saying though, I would like to see APUs in a thinner form factor laptop.
Posted on Reply
#39
newtekie1
Semi-Retired Folder
AquinusDoesn't Intel own the trademark of "Ultrabook" so don't they have a certain level of control over what devices can be called Ultrabooks?
I know what you're saying though, I would like to see APUs in a thinner form factor laptop.
That is why I said "Ultrabook form factor". Intel owns the name, but there is nothing stopping manufacturers from putting AMD hardware in laptops that meet the same form factor. They just can't call them Ultrabooks.
Posted on Reply
#40
HumanSmoke
btarunrAgain, what's new with Richland:
  • Higher CPU and GPU clock speeds
  • Higher optimal memory speed (DDR3-2133 MHz)
  • GPU gets GCN architecture
.
FourstaffTrinity wasn't GCN.
What's with all the GCN talk ? Afaia, Richland iGPU is a tweaked previous gen VLIW4
Posted on Reply
#41
newtekie1
Semi-Retired Folder
HumanSmokeWhat's with all the GCN talk ? Afaia, Richland iGPU is a tweaked previous gen VLIW4
Nope, read the first post again, Richland has an actual upgraded GPU that uses GCN.
Posted on Reply
#42
HumanSmoke
newtekie1Nope, read the first post again
Cant I just use some common sense and AMD's own literature from CESearlier this month?
Pick the odd one out...You'd think that if Richland was GCN, AMD wouldn't have made the distinction between GCN and "2nd Generation DirectX11 GPU"
newtekie1Richland has an actual upgraded GPU that uses GCN.
You want to bet cash money on that ?
Posted on Reply
#43
NeoXF
Whatever the case may be, I'm willing to bet they will bring a great bang for the buck and give a good squeeze to Intel's lower end parts untill Steamroller/Kaveri comes.


Also, can someone remind me, are AMD module-based CPUs able to clock only one cores from inside that module? I remember so...
Posted on Reply
#44
jagd
Is this roadmap slide new or old ?
Posted on Reply
#45
Aquinus
Resident Wat-man
NeoXFAlso, can someone remind me, are AMD module-based CPUs able to clock only one cores from inside that module? I remember so...
That would defeat the purpose of having a module...
HumanSmokePick the odd one out...You'd think that if Richland was GCN, AMD wouldn't have made the distinction between GCN and "2nd Generation DirectX11 GPU"
Well considering Trinity is blank and everything else in the same time frame is saying "DX11 Capable GPU" not "2nd gen DX11 GPU," so it does imply that Richland might be getting something different. It might not be GCN, but it doesn't seem to be VLIW5 either.
Posted on Reply
#46
HumanSmoke
jagdIs this roadmap slide new or old ?
If you’d care to read my post I think it will answer your question:
HumanSmokeCant I just use some common sense and AMD's own literature from CES earlier this month?
CES was held in Las Vegas three weeks ago (8-11 January)- although I do take your point. Being three weeks old probably means that AMD have a new roadmap/ powerpoint slide presentation in place by now.
AquinusWell considering Trinity is blank and everything else in the same time frame is saying "DX11 Capable GPU" not "2nd gen DX11 GPU," so it does imply that Richland might be getting something different. It might not be GCN, but it doesn't seem to be VLIW5 either.
:shadedshu
I don’t remember saying that Richland is VLIW5…maybe because I didn’t. If you’d have actually read my post (#41) it says:
HumanSmokeWhat's with all the GCN talk ? Afaia, Richland iGPU is a tweaked previous gen VLIW4
1st gen DX11 GPU = VLIW5 = Cypress et al
2nd gen DX11 GPU= VLIW4 = Cayman et al
Posted on Reply
#47
Aquinus
Resident Wat-man
HumanSmoke:shadedshu
I don’t remember saying that Richland is VLIW5…maybe because I didn’t. If you’d have actually read my post (#41) it says:
Wow, maybe it's because I was reading what the image you posted was saying and not what you wrote. No need to get defensive about it. If DX11 gen 1 is VLIW4 then it would be a good guess that gen 2 is VLIW5. You don't need to say it for it to make sense... Get off your high horse and calm down. You've obviously getting way more worked up about this than anyone should. This thread will still be here after you take a couple deep breaths.

I also take "road maps" with a grain of salt. We really won't know until there are verified engineering samples of the CPU.

All in all, it might be GCN then again it might not. We don't know.

Let's wait and see!
I can think of a number of things to do between now and then. :p
Posted on Reply
#49
Aquinus
Resident Wat-man
xorbeQuoted for fail ... and it's old, old information.
www.lmgtfy.com/?q=vliw5+vs+vliw4
The only thing that is fail here is the results of that google search. I was talking about the image, not the instruction set itself. I'm not quite sure what you're trying to get at here. So instead of saying that what I said is "fail" lets start by saying why, otherwise you're just trolling.
Posted on Reply
#50
HumanSmoke
AquinusWow, maybe it's because I was reading what the image you posted was saying and not what you wrote. No need to get defensive about it...Get off your high horse and calm down...take a couple deep breaths.
Really? You managed to infer quite a bit didn't you. Pity it was off the mark. Aside from a tongue-in-cheek jab at the propensity that AMD has for PPS/timetable revisions, I don't see anything defensive or supercilious about my post. But if you're willing to ascribe motivation to myself, maybe I should return the favour...so, if anything you're the one getting defensive-You've just noted that you don't know what µarch Richland is...yet blast me for using an educated guess, and AMD's own literature.
AquinusIf DX11 gen 1 is VLIW4 then it would be a good guess that gen 2 is VLIW5. You don't need to say it for it to make sense
Actually, it should be the other way around :rolleyes: - and probably what xorbe was attempting to convey to you. VLIW5 predates VLIW4 in DX11 capable GPUs.
The VLIW5 HD 5870 (Cypress GPU) launched 23 Sept, 2009 was the first DX11 card. The VLIW4 HD 6970 (Cayman GPU) launched 15 Dec., 2010.
Posted on Reply
Add your own comment
Apr 19th, 2024 22:18 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts