Welcome Guest ( Log In | Register )

Outline · [ Standard ] · Linear+

 AMD Bulldozer & Bobcat

views
     
lex
post Oct 24 2011, 09:36 PM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
QUOTE(banks @ Oct 24 2011, 05:42 PM)
FX-8150 looks pretty decent on Linux benchmarks
Already posted a few earlier, like here (FX-8150 vs Core i7 2600K, Phenom II X6 1090T, Core 2 Quad Q9550, Athlon X2 6000+, etc) and here (FX4100 vs Fusion vs Sandy Bridge). There are also a number of new ones: OpenBenchmarking - AMD FX-8150 vs. Intel On Linux, OpenBenchmarking - AMD FX-8150 vs. Intel On Linux, OpenBenchmarking - Bulldozer, Gulftown, Sandy Bridge, Shanghai Benchmarks, OpenBenchmarking - Bulldozer, Gulftown, Sandy Bridge Scaling, etc (more coming). wink.gif

Video reviews from Motherboards.org (the packaging guy)...


Video reviews from Linus Tech Tips...


Video reviews from PCWizKids Tech Talk.com (newcomer?)...


Latest reviews here: PC Watch > Top AMD FX series "FX-8150" Benchmark (from Japan) and TweakPC.de > AMD's Bulldozer FX-8150 (has dual Radeon HD6970 CrossFireX and Eyefinity benchmarks). icon_rolleyes.gif

This post has been edited by lex: Oct 24 2011, 09:38 PM
lex
post Oct 27 2011, 09:08 AM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
QUOTE(chenwah88 @ Oct 26 2011, 08:18 AM)
http://www.agner.org/optimize/blog/read.php?i=49
Is it all benchmarks were crippled amd?
Majority of software do not use ICC (Intel's C Compiler, used in specialized server, HPC and custom applications). Take for example Microsoft products (compiled with Microsoft's own compilers), or even compare the results compiled with GCC (GNU C Compiler) under Linux in the previous posts. wink.gif

QUOTE(dstl1128 @ Oct 27 2011, 08:25 AM)
What I mean was the result is 'different' from other review sites Cinebench 11.5 benchmark result.
You should check the review again, because they disabled the Turbo functions to check and compare IPC differences between Bulldozer, Thuban and Sandy Bridge. wink.gif
lex
post Oct 27 2011, 09:26 PM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
Latest review here: Tom's Hardware > All Reviews > Components > CPU > AMD FX: Energy Efficiency Compared To Eight Other CPUs with Bulldozer versus Deneb, Thuban, Bloomfield, Lynnfield, Gulftown and Sandy Bridge. Another earlier (very short) review here: Digital Versus: AMD FX-8150. wink.gif

Related review here: Legit Reviews - ASUS Crosshair V Formula 'Bulldozer' Motherboard Review pits Phenom II X6 1100T against FX-8150, including CrossFireX and SLI performance. Also another article on Bulldozer from ExtremeTech here: ExtremeTech > Computing > AMD may be betting future success on Bobcat, not Bulldozer.icon_rolleyes.gif
lex
post Nov 5 2011, 09:07 PM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
QUOTE(saturn85 @ Nov 2 2011, 05:00 AM)
This record is broken again: AnandTech > CPUs > Bulldozer Does It Again: Overclocked to 8.58GHz...
QUOTE
Bulldozer Does It Again: Overclocked to 8.58GHz

by Kristian Vättö on 11/3/2011 7:45:00 AM
Posted in CPUs, AMD, Bulldozer, FX-8150


user posted image

Last week, we reported that Andre Yang had managed to overclock AMD's FX-8150 CPU to 8.46GHz, breaking the former record set by AMD. Andre Yang has now been able to break his own record by roughly 120MHz, which is fairly notable increase given that the previous record he set was only 30MHz greater than the record before that. The new world record is 8584.8MHz to be exact, 123.3MHz greater than the previous. The record is not official yet, though, as it has not appeared in CPU-Z's validation database.

user posted image

Andre used the same Crosshair V Formula motherboard and the only difference seems to be the higher core voltage, which was 2.076V in the new record (1.992V in the old one). Andre used liquid nitrogen for cooling, so even higher overclocks could be achieved with liquid helium, which liquefies at -269 Celsius while it's -196 Celsius for nitrogen in atmosphere pressure. Bulldozer's upcoming B3 stepping could also improve the overclock-ability, possibly making 9GHz plausible.

Source: XtremeSystems


QUOTE(kingkingyyk @ Nov 5 2011, 04:16 PM)
Saw Trinity engineering sample benchmark from chiphell.
3.8GHz, 4.1GHz turbo boost.
I believe it is 125W SKU.
The CPU uses Piledriver architecture.
Benchmarked in Cinebench, on par with current Llano only. doh.gif
How to compete to Ivy Bridge i3? No way.
AMD lost another card to play with.
You must be referring to this: CHIPHELL > Forum > Computer > CPU / Memory / Motherboard / Overclocking > Trinity - 3 big SB together. ...
QUOTE(bigpao007 @ Nov 4 2011, 19:39 PM)
[CPU] Trinity - 3 big SB together.

If you want to see in this the conclusion: Trinity Trinity is three big SB together.

This article is only one pair of AMD, said the anger, love the voice of DIY players.

First to a currently known about the characteristics of Trinity's architecture diagram:

user posted image
The figure is clearly expressed:

Trinity is the use of second-generation Bulldozer core and the core of NGC shows, the interface for the FM2.

CPU performance improved relative to the first generation of APU Llano up to 30%; display up to 30% performance increase.

Okay, so take a look at why I say it is 3 SB together.

First SB: interface is not compatible

Without further ado, the last picture shows the card.
user posted image

The figure above on the left of the interface of the first generation FM1 APU Llano, on the right interface for the FM2 second generation APU Trinity.

We see that different? ?

Seen it?

Out of it?

FM1 is 905 PIN, FM2 for the 904 PIN,

In fact, this is not the key to the

The following key FM1 FM2 CPU and motherboard comparison of the interface.

user posted image

Dear AMD FANS, do you think FM2 FM1 of the CPU board can be plugged into it? ?

You can plug in to it?

You can go up it?

Can it?

Second SB: Power

user posted image

FM1 is only 65/100, FM2 it? ?

More out of a 125W, but the number of cores it? ?

Only four are still the biggest heart.


According to the current first generation APU Llano performance, even if the second-generation Bulldozer core than the first generation APU Llano increased by 30% (20% of this is AMD's own targets on the slide), do you think the top of the Trinity of the CPU performance can be

SNB I3 can win it?

Could win it?

Could it?

Possible?

Timing was not released even when the face of the Trinity is IVY version of the I3, more robust performance IVY I3 of.

Third SB: Frequency

Here is the ES SAMPLE naming convention, I will focus on the ES version is released the CPU frequency.

user posted image
user posted image
user posted image

See that? Highest 3.8G, can TC to 4.1

The first generation of APU Llano frequency than TC 0.3, TC个G,


And such a high frequency, but the actual performance is almost the same and Llano (LZ tested 3D MARK V and R11.5, but I am afraid I can not release the current picture)

LZ at this time speechless.

Are there any players do not know what to say: BIOS, system drivers to improve performance ah.

I suggest that you search I found the first half of the release of the first generation of bulldozers R11.5 and 3D MARK06 CPU scores, to contrast the present and see what there is to enhance ......

As a love of DIY products and also love the players AMD CPU, AMD once again deeply hurt you.
It also seems that FM2 is not compatible with FM1. hmm.gif

QUOTE(flexus90 @ Nov 5 2011, 04:18 PM)
Is it official from AMD? Unheard this before.
That's old news, but its official and in limited quantities. wink.gif

Latest reviews: AMD FX performance in multi-GPU setups here: HARDOCP - Introduction - AMD FX-8150 Multi-GPU Gameplay Performance Review and [Madshrimps] AMD FX 8150 Revisited. icon_rolleyes.gif

This post has been edited by lex: Nov 5 2011, 09:11 PM
lex
post Nov 23 2011, 10:18 PM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
QUOTE(Thrust @ Nov 23 2011, 12:10 PM)
Actually under Linux OS, The FX8150 is on par & even slightly faster than the i7 990X base on IPC. Well that's what I saw on Phoronix review.

I seems Windows 7 isn't optimized for Bulldozer sad.gif
Not quite on par overall, as the Core i7 990X was never a competitor to FX8150 (even the pricing is a telling sign). For example, you can check the results here: OpenBenchmarking.org - AMD FX-8150 Vs. Intel On Linux Benchmarks (quite a few others at OpenBenchmarking.org as well). wink.gif

QUOTE(Najmods @ Nov 23 2011, 05:42 PM)
Well, the thread scheduling is done so to trigger higher Turbo Core clock because of it uses fewer modules. So it could simply be done by overclocking manually anyway.

But I like this quote in particular near the end of the article, it could be beneficial to AMD somewhat
Bulldozer cores are really more like hardware threads since CMT is another type of multi-threading technology (like SMT or HyperThreading). It also has a few niggling issues (such as drop in performance) as found in HyperThreading with certain software or scenarios. Can be seen that in this new review: AnandTech - Bulldozer for Servers: Testing AMD's "Interlagos" Opteron 6200 Series (look at Rendering Performance: 3DSMax 2012 and Blender results on this page: Rendering Performance: Maxwell Render and Blender). And yups, Interlagos reviewed at last. hmm.gif

lex
post Jan 20 2012, 06:25 AM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
News: VR-Zone.com - Microsoft Re-Releases Patches for AMD Bulldozer Architecture...
QUOTE
VR-Zone.com - Microsoft Re-Releases Patches for AMD Bulldozer Architecture
Reported by Theo Valich on Friday, January 13 2012 2:57 am

Few weeks back, Microsoft caused a lot of fuss by releasing an incomplete Windows 7 / Server 2008 R2 patch for AMD Bulldozer architecture, which should increase the performance. Download those today.

user posted image

NEWS

Several weeks ago, Microsoft caused quite a confusion by releasing a patch that was supposed to increase performance of AMD Bulldozer architecture (FX and Opteron 6200 processors) and its Windows 7 and Windows Server 2008 R2 operating systems.

If you recall, the task scheduler in upcoming Windows 8 operating system properly detects and executes threads on the AMD Bulldozer architecture, while Windows 7 based operating systems do not. However, Microsoft recalled the patch as the release was incomplete and could have caused system instability.

Just as AMD elaborated, there are two patches for the architecture, and they need to be installed in the correct order.

If you own a Bulldozer based system, you have to install in this order:

FIRST INSTALL THIS KB2645594
The CPU scheduling techniques that are used by Windows 7 and Windows Server 2008 R2 are not optimized for the AMD Bulldozer module architecture. Therefore, multithreaded workloads may not be optimally distributed on computers that have one of these processors installed in a lightly-threaded environment. This may result in decreased system performance for some applications. When this update is installed, the scheduler will be aware that your Bulldozer processor contains dual-core modules. In essence, threads 1-4 now get assigned to their own module first.

THEN INSTALL KB2646060
The CPU Power Policies that are used by Windows 7 and Windows Server 2008 R2 are not optimized for the dual core AMD Bulldozer module architecture. This can result in decreased system performance with multithreaded workloads in lightly-threaded environments. When this update is installed, Bulldozer modules will be less likely to achieve the C6 power state. This potentially results in increased power consumption in more lightly-threaded environments.


Let us know in your comments if this patch actually increased the performance of your system or not.


News: ExtremeTech: Microsoft patches Bulldozer’s performance — we investigate...
QUOTE
Microsoft patches Bulldozer’s performance — we investigate
By Joel Hruska on January 12, 2012 at 4:29 pm

user posted image

When AMD launched Bulldozer back in October, the company claimed that problems with Windows 7 thread scheduler prevented the CPU from delivering peak performance. According to AMD, Windows 7 didn’t accurately understand how best to schedule threads to take advantage of the company’s shared architecture, and as a result “there are possibilities where opportunities for resource sharing or activate [sic] higher Turbo Core frequencies are missed.”

Microsoft has just released a pair of hotfixes that claim to resolve the issues that handicapped Bulldozer. Curious to see what might have changed, we dusted off our test rig, installed both updates, and ran some of the same tests we’d previously checked.

The updates in question are KB2646060 and KB2645594. The first changes how often Bulldozer achieves the C6 power state. MS notes that “this potentially results in increased power consumption in more lightly-threaded environments.” The second update is more vague, mentioning that “multithreaded workloads may not be optimally distributed… in a lightly-threaded environment. This may result in decreased system performance for some applications.”

user posted image

AMD’s own statements on how much the update improves performance are similarly modest. “Our testing shows that not every application realizes a performance boost. In fact, heavily threaded apps (those designed to use all 8 cores), get little or no uplift from this hotfix – they are already maxing out the processor. In other cases, the uplift averages out to a 1-2 percent uplift.” writes Adam Kozak, AMD’s product marketing manager.

Here’s what we saw when we installed the updates and started testing. As in our original article, Turbo Core was disabled. A 4M/8C configuration means that a test was run on all four modules with both cores enabled per module. Previously, a 4M/4C configuration was notably faster than 4M/8C in quad-threaded workloads do to scheduling problems.

user posted image

user posted image

The good news is that these updates improve Bulldozer’s performance by 2.74% in Cinebench and 3.33% in Maxwell Render. The bad news is just how limited the updates are. As AMD notes, these boosts are limited to “lightly threaded environments” — we saw no performance improvement when we benchmarked these tests with all eight cores enabled.

Hopefully the updated scheduler and AMD’s own admission that it’s of limited use will help put a finish to persistent rumors that reviewers, Intel, and the Illuminati have somehow conspired to make Bulldozer’s performance out to be worse than it actually is. AMD’s demos at CES 2012 have focused nearly exclusively on the company’s graphics performance, as evidenced by this screenshot of the company’s CES page:

user posted image

The demos of up-and-running Trinity silicon confirmed reports that the chip is ramping well and that GPU performance is a substantial improvement over Llano, but too many websites have combined two separate statements from AMD. First, that Trinity is based on Bulldozer and has up to four cores, and second, that AMD is targeting a 17W power envelope for its own ultra thin-and-light segment.

The fact that both of these statements are true doesn’t mean that AMD is planning to launch a 17W quad-core based on Piledriver. Even Intel never tried to hit anything near that with a mainstream Sandy Bridge part, opting instead to wait for 22nm and the debut of Ivy Bridge.

As we head towards AMD’s Financial Analyst Day in February, everything is rosy on the graphics side — at least, apart from the company’s inability to earn any significant profit from the business — but the CPU side of the equation is full of questions. Bulldozer isn’t getting any magic software-side improvements. Trinity’s GPU is awesomesauce, but the CPU will be doing well to match Llano’s performance-per-watt and 28nm follow-ups to Bobcat remain conspicuous in their absence.


This post has been edited by lex: Jan 20 2012, 06:27 AM
lex
post May 31 2012, 05:25 AM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
AnandTech analyzes Bulldozer's IPC: AnandTech > IT Computing > The Bulldozer Aftermath: Delving Even Deeper. wink.gif
lex
post Jun 4 2012, 06:05 PM

Old Am I?
Group Icon
VIP
18,182 posts

Joined: Jan 2005
From: Dagobah
QUOTE(tplus1 @ Jun 4 2012, 10:30 AM)
So it is not recommended to get the fx 8 cores cpu?
The new FX8000 series isn't exactly true eight core, more like AMD's version of HyperThreading implemented in hardware. In fact its closest competitor in terms of performance is the older Phenom II X6 1100T. Its not that recommended given the pricing (still in the RM900 range), less-than-stellar performance and high power consumption. When compared with competitor's Core i7 2600K or Core i7 3770K, the FX8150 loses almost/nearly all the benchmarks (sometimes by significant margins in certain applications, most games and in multi-GPU setups). hmm.gif

QUOTE(tplus1 @ Jun 4 2012, 10:30 AM)
All review i read is talking about how i5 can outperform fx8120 or fx8150 easily. Even the sandy bridge i5 2500k?
Yes, mostly in applications that are not heavily multi-threaded (plenty of those in everyday usage) and especially games. Majority of games are not highly multi-threaded, prefers high IPC (instructions per clock) and good per-core performance. wink.gif

This post has been edited by lex: Jun 4 2012, 06:07 PM

4 Pages « < 2 3 4Top
 

Change to:
| Lo-Fi Version
0.0401sec    0.56    7 queries    GZIP Disabled
Time is now: 26th November 2025 - 08:51 PM