Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Combat Pretzel
Jun 23, 2004

No, seriously... what kurds?!
That doesn't look anything like other AMD presentations.

Anyway, from how I understand it, a BD module is essentially two integer cores and one FP core with a common decoder frontend. So in regards to integer performance, the BD should blow the SB out of the water. Since it'd be two integer pipelines versus one hyperthreaded one (on the 2600K).

And in regards to FP, I think the BD pipeline was double width and was supposed to do parallel work for both submodules with 128bit SIMD instructions? I don't think the Intel FP 256bit pipeline can internally parallelize stuff with multiple independent 128bit SIMD instructions. So again, performance should be close. Only with AVX there should be a discernable difference, since BD can either supply a whole set of 128bit registers for each module, or a set of 256bit one for one submodule.

Then again, some developers of "math" heavy code don't seem to dig SSE, AVX and the like, because they're float only. So it wouldn't mean much, anyway. For instance, the x264 developer thinks that it's a pile of poo poo (it's mostly integer math).

Adbot
ADBOT LOVES YOU

freeforumuser
Aug 11, 2007

Factory Factory posted:

Is that the top 4-core or comparing 8-core to the i7-2600K's 4-core? AMD seems to be going all-in on multithreaded performance if that's an 8 vs. 4 comparison. Then again, that would mirror their graphics strategies.

It'll kinda suck when 6/8-core Sandy Bridge E parts come out. Poor AMD :saddowns:

Most likely the Zambezi 8 core / 4 module part. If the chart is true I can understand why Intel was selling SB on the cheap.

Still, we don't know how the frequency, overclocking and power consumption would work out.

PC LOAD LETTER
May 23, 2005
WTF?!

tijag posted:

then what you're seeing is effectively the performance of a 2 module 4 core BD chip being equivalent to a 2600K.
Its supposed to be the 8 "core" 4 module chip that is being bench marked there. Clock speed is unknown but its probably at or near peak base speed at the very least.

Combat Pretzel posted:

So in regards to integer performance, the BD should blow the SB out of the water.
AMD's module approach is more hardware intensive than Intel's hyperthreading but yeah its also supposed to get you more lots performance. Its probably the next best thing to going all out and putting a full independent core on die.

PC LOAD LETTER fucked around with this message at 05:45 on May 5, 2011

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.
If it is real, I will be :stare::fh: all over that poo poo.

tijag
Aug 6, 2002

PC LOAD LETTER posted:

Its supposed to be the 8 "core" 4 module chip that is being bench marked there. Clock speed is unknown but its probably at or near peak base speed at the very least.

I'm aware of that, but if the benchmark only effectively uses 4 cores, then the performance you are seeing is closer to that, or nearly identical to, a 4 core BD chip.

it also makes the 2600K similar or almost identical in performance to a 2500K.

If thats the case, AMD has a decent chip, because being able to compete with the 2500K with a quad core BD would be amazing. Current AMD offerings are probably 30% slower than a 2500K.

spasticColon
Sep 22, 2004

In loving memory of Donald Pleasance
Now that I already pulled the trigger on a 2500K setup I'm sure AMD will now make their big comeback. :negative:

You Am I
May 20, 2001

Me @ your poasting

spasticColon posted:

Now that I already pulled the trigger on a 2500K setup I'm sure AMD will now make their big comeback. :negative:

AMD will find someway of screwing it up, like dodgey third party chipsets or some CPU bug.

PC LOAD LETTER
May 23, 2005
WTF?!
I don't think there are going to be any 3rd party chipsets for AMD's chips this time around. Their in house chipsets aren't perfect but do seem to be good enough that no one tries to compete with them anymore.

They could certainly still screw something up though.

edit: looks like some specs for some up coming AMD 900 series based motherboards have been leaked as well. No prices but I'm sure you could make a reasonably close WAG on what they'll be based on ASUS's current line up.

PC LOAD LETTER fucked around with this message at 02:05 on May 6, 2011

Cuntpunch
Oct 3, 2003

A monkey in a long line of kings

You Am I posted:

AMD will find someway of screwing it up, like dodgey third party chipsets or some CPU bug.

My guess would be a failure to compete in price/performance, or a failure to compete in overclocking - both would be strange reversals for AMD so I hope I'm wrong! My guess is just because the 2600k as a target for a 'top end' chip is deceptive because it's reasonably cheap for Intel, and also overclocks like crazy.

freeforumuser
Aug 11, 2007

You Am I posted:

AMD will find someway of screwing it up, like dodgey third party chipsets or some CPU bug.

Not going to happen since AMD completely ditched 3rd party chipsets 3 years ago. Quality has only shot up tremendously compared to the days of quirky nForce 3/4 (but at least that was still much better than Nvidia disastrous 680/780/790i chipsets)

eames
May 9, 2009

You Am I posted:

AMD will find someway of screwing it up, like dodgey third party chipsets or some CPU bug.

Then it’ll probably be their last screwup, because I don’t see how they have a chance in the x86 game once Ivy Bridge is out.
See Intel’s 22nm Tri-Gate Transistors and AMD’s lack of having something comparable within the near future.

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.

eames posted:

Then it’ll probably be their last screwup, because I don’t see how they have a chance in the x86 game once Ivy Bridge is out.
See Intel’s 22nm Tri-Gate Transistors and AMD’s lack of having something comparable within the near future.

Mmm, Kool-Aid.

Ryokurin
Jul 14, 2001

Wanna Die?

eames posted:

Then it’ll probably be their last screwup, because I don’t see how they have a chance in the x86 game once Ivy Bridge is out.
See Intel’s 22nm Tri-Gate Transistors and AMD’s lack of having something comparable within the near future.

FinFETs have been around for a while and wasn't invented by Intel so the only thing they can block is trade secrets on how they design chips around their use. TSMC also demoed similar technology around the same time Intel did a decade ago and have said they will start to use it at 20nm. The big hold out is Global Foundries which hasn't really said much about their finFET plans at all.

Then there's also the statement of one of the developers of finFETs saying that they won't result in better performance once we go below 22nm anyways. I think the key is going to be if Intel's process is to allow a performance difference sub 22nm or if they are pushing it just to lower power to better compete with ARM.

PC LOAD LETTER
May 23, 2005
WTF?!
Looking at how Intel configured SB (only ~10% performance improvement over Nehalem but around 20% less power usage) my WAG is they're going to emphasize power usage over performance. They already know good and well that most of their CPU's are overpowered for what most people need to do with their computers and they're going to have to compete with ARM in laptops and servers somehow.

AMD's, well GF's fab tech is based off of what IBM has IIRC, which is actually part of a whole group of companies working together to share development costs of the process. I don't know what they're planning re: FinFET's either.

edit: Kanter at RWT recently did a article on Intel's FinFET's you guys may be interested in. One of the relevant tidbits in it was that GF/IBM were planning on using FinFET's when they get to a 14nm process. They appear to be shooting for a late 2015 launch for that process. I somehow doubt that they'll be able to avoid some delays or even pull that date in earlier so I think it wouldn't be unreasonable to expect a 14nm chip using FinFET's from AMD until early 2016 at best.

It'll be interesting to say the least to see how AMD tries to compete when Intel will have such a huge advantage for so long. One option would be to use FD-SOI I guess since that may be available if not now then much sooner than 2015, but Kanter mentions that would increase costs by 10%. AMD has gone the SOI route before though when others wouldn't due to reported cost issues, so we'll have to wait and see...

PC LOAD LETTER fucked around with this message at 01:39 on May 9, 2011

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.
Any new developments in AMDland?

PC LOAD LETTER
May 23, 2005
WTF?!
Not really no, just some more false benches to troll people. Some mobo manufacturers did mention they'd show some 9xx based mobos at Computex.

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
Nordic Hardware has a rundown of the Llano benchmarks leaked by DonahimHaber. They could of course be BS, but they seem pretty reasonable as benches of IGP game performance versus Sandy Bridge. It still won't play Crysis, but everything else works well. Fudzilla also has details on the top of the Llano mobile lineup. The highest-end mobile CPU will be the Fusion A8-3530MX, a 32nm quad-core with a base clock of 1.9GHz, Turboing up to 2.6Ghz. It features Radeon HD 6620G graphics (400 shaders @ 444Mhz). This compares to the current Phenom II X4 940BE, a 2.4Ghz 45W CPU without graphics.

Like I've said in the past, the critical aspect is how well AMD's Turbo Core works. The graphics seem pretty compelling, so per-thread CPU performance is the only real question.

PC LOAD LETTER
May 23, 2005
WTF?!
Llano mobo details if that is your thing. Looks decent enough even if it is a ECS mobo but no real surprises yet.

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.
I know this is pretty :downs: of me, but is BD going on laptops, or is it just Llano?

Bob Morales
Aug 18, 2006


Just wear the fucking mask, Bob

I don't care how many people I probably infected with COVID-19 while refusing to wear a mask, my comfort is far more important than the health and safety of everyone around me!

Sinestro posted:

I know this is pretty :downs: of me, but is BD going on laptops, or is it just Llano?

Has AMD ever had a compelling laptop chipset? (Except for the current Zacate setups)

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.

Bob Morales posted:

Has AMD ever had a compelling laptop chipset? (Except for the current Zacate setups)

The initial platform kicked the Pentium 4-M's rear end, Puma was better than Intel's ULV offering, Champlain is only as bad as AMD's current desktop offering, and Kite Enhanced invented the new SSD cache tech in Z68.

EDIT: Me english real good.
EDIT 2: 4-M != M.

Sinestro fucked around with this message at 16:59 on May 19, 2011

HalloKitty
Sep 30, 2005

Adjust the bass and let the Alpine blast

Bob Morales posted:

Has AMD ever had a compelling laptop chipset? (Except for the current Zacate setups)

Possibly only around the time Intel was pushing Pentium 4-M, because everyone knows Netburst in a laptop was a great idea

PC LOAD LETTER
May 23, 2005
WTF?!

Sinestro posted:

I know this is pretty :downs: of me, but is BD going on laptops, or is it just Llano?

I'm sure eventually it'll go into laptops but I don't know when. The June launch is just for desktop and server parts AFAIK.

e: more mobo shots.

ASUS high end 990FX mobo.
Jetway "cheap" 990X mobo.

Prices on both unknown but I'd expect the ASUS board to be close to or north of $200, the Jetway board looks like a mid range offering which is uncommon from them.

PC LOAD LETTER fucked around with this message at 13:39 on May 19, 2011

roadhead
Dec 25, 2001

Sinestro posted:

I know this is pretty :downs: of me, but is BD going on laptops, or is it just Llano?

Last I heard the mobile equivalent of Bulldozer is Bobcat, but I might be mistaken there.

And by equivalent I mean you would need discrete graphics to use Bobcat. Does anyone know more?


EDIT: Don't listen to me I'm stupid. Bobcat is just the re-designed CPU core for the upcoming "below Llano" notebooks. I don't think they are going to do many "new" designs for notebooks that aren't APU - so high-end notebook workstations will just continue to adapt the HE desktop chips maybe?

roadhead fucked around with this message at 15:12 on May 19, 2011

ultrabay2000
Jan 1, 2010


Sinestro posted:

The initial platform kicked the Pentium M's rear end

uh what.

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.

SaderBiscut posted:

uh what.

I meant 4-M. As HalloKitty said, mobile Netburst = :downsbravo:.

Sinestro fucked around with this message at 16:20 on May 19, 2011

fishmech
Jul 16, 2006

by VideoGames
Salad Prong
Well yeah but beating "mobile" Pentium 4s is like beating a guy with a broken leg in a race. :v:

ultrabay2000
Jan 1, 2010


My T42 completely blew my A31 out of the water despite having a lower clock-speed and the same amount of ram, but you know.

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.

SaderBiscut posted:

My T42 completely blew my A31 out of the water despite having a lower clock-speed and the same amount of ram, but you know.

That is Pentium M, which became Core. This is the Pentium 4-M/ Mobile Pentium 4, which is just a Northwood P4 (later Prescott) with the TDP scaled down.

wicka
Jun 28, 2007


What is Zacate?

Sinestro
Oct 31, 2010

The perfect day needs the perfect set of wheels.

wicka posted:

What is Zacate?

Two Bobcat cores (distant decendant of Clawhammer optimised for low TDP) with a IGP on die.

TacticalHoodie
May 7, 2007

freeforumuser posted:

Not going to happen since AMD completely ditched 3rd party chipsets 3 years ago. Quality has only shot up tremendously compared to the days of quirky nForce 3/4 (but at least that was still much better than Nvidia disastrous 680/780/790i chipsets)

The Via chipsets were even worse off for AMD. Hell, when I bought my Althon 64 3400+ to replace a brunt out Pentium 4, I had to choose from the Via Chipset whose drivers degraded system performance upon each new release or the nforce chipset that flat out refused to boot many AGP video cards at the time, including the BFG 6800 GT I had at the time. I was happy to get my e8400 and intel-based chipset motherboard after that cluster-gently caress VIA/AMD combo.

Now I know AMD is using a first party chip set makes me really want to go with them again if the benchmarks and pricing is good. Even if the performance is a few notches below the i5 2500, having SLI motherboards and the potential price advantage might make me go to AMD.

Unless they gently caress up like Intel did with the P67 or the vendors make piss-poor boards.

dad on the rag
Apr 25, 2010
Possible Llano and Bulldozer prices have leaked. FX 8 cores from $290-350, 6 cores for $240, and 4 cores for $190.

Could somebody translate it through google and post the text? For some reason reason I can't get it to work in Opera/Firefox.

dad on the rag fucked around with this message at 22:19 on May 20, 2011

Alereon
Feb 6, 2004

Dehumanize yourself and face to Trumpshed
College Slice
Here's a Google translation of the Donanim Haber article with the original image attached.

Google Translate posted:

Bulldozer FX and AMD's next-generation Fusion processors LIano series began to arrive on the price information. Upcoming launch date with each passing day, new details emerged the AMD front, not yet officially doğrulanmamakla with four, six and eight-core processors are a new generation of end-user retail prices published by a Chinese site. News from the table in order to be understood according to the model of parallel processors 160-400 will have integrated graphics unit, code-named Fusion LIano A series of processors will be presented suggests the level of $ 70-170. As is known, these processors are based on Husky K10.5 core architecture design using the updated version.

AMD's high-performance segment konumlandıracağı an aggressive pricing strategy Bulldozer FX processors will come in as expected. 32nm process technology and according to the model until 8MB'a 8MB'a L2 and L3 memory capacity of up to set prices for its new generation Bulldozer FX processor is varied between $ 190 and $ 320. AMD's official statements to be remembered Bulldozer FX 8-core processor, Intel's Core i7-2600 karşılaştırılyordu model. The resulting price chart and the current selling price of the Core i7-2600K'nın also taken into account, AMD's fastest processor 8-core Core i7-2600 also are growing, with estimates that will fight.

Only registered members can see post attachments!

BlackMK4
Aug 23, 2006

wat.
Megamarm
Edit: redacted

unpronounceable
Apr 4, 2010

You mean we still have another game to go through?!
Fallen Rib

Alereon posted:



They are really specific with the TDP's there on the 6 and 4 core FX chips. Now that I look at it, can anyone guess why the 4 core TDP > 6 core TDP > 8 core TDP?

Factory Factory
Mar 19, 2010

This is what
Arcane Velocity was like.
Higher clock speeds on the lower-core processors, maybe?

wicka
Jun 28, 2007


I would definitely be willing to spend $190 on a quad core.

spasticColon
Sep 22, 2004

In loving memory of Donald Pleasance
Any word yet on the overclocking ability of these chips?

Adbot
ADBOT LOVES YOU

jassi007
Aug 9, 2006

mmmmm.. burger...

LiftAuff posted:

Possible Llano and Bulldozer prices have leaked. FX 8 cores from $290-350, 6 cores for $240, and 4 cores for $190.

Could somebody translate it through google and post the text? For some reason reason I can't get it to work in Opera/Firefox.

Another site I read said that was the per tray price, not retail msrp.

  • Locked thread