Sign in with
Sign up | Sign in

Enabling Turbo Core

AMD FX-8150 Review: From Bulldozer To Zambezi To FX
By

Back when AMD launched its Phenom II X6 1090T, it debuted a feature called Turbo Core. It was supposed to be an answer to Intel’s Turbo Boost capability, which would capitalize on available TDP in poorly-threaded workloads (where other cores simply sat idle) to increase clock rate.

As you know, Turbo Boost (Intel’s feature) employs an on-die power controller that evaluates temperature, current, power consumption, and operating system states. With all of that information, it can shut down idle cores, freeing up thermal headroom to accelerate active cores. The degree of acceleration is contingent on how many cores are in use. Obviously, there’s a lot more room to ratchet up clock rate in a single-threaded application. As a result, you end up with a frequency map of sorts that scales up and down depending on the parallelism of any given application. As a quick example, from Intel Core i5 And Core i7: Intel’s Mainstream Magnum Opus:

Turbo Boost: Available Bins (Under TDP/A/Temp)
Processor Number
Frequency
4 Cores Active
3 Cores Active
2 Cores Active
1 Core Active
Core i7-870
2.93 GHz
2
2
4
5
Core i7-860
2.8 GHz
1
1
4
5
Core i5-750
2.66 GHz
1
1
4
4
Core i7-975
3.33 GHz
1
1
1
2
Core i7-950
3.06 GHz
1
1
1
2
Core i7-920
2.66 GHz
1
1
1
2


Turbo Core (AMD’s feature), in comparison, was presented as a deterministic feature that turns on in lightly-threaded workloads where three or fewer cores are active, or off altogether when an application taxes anything more than three cores. In practice, it didn’t seem nearly as binary as AMD described. What I saw in AMD Phenom II X6 1090T And 890FX Platform Review: Hello, Leo was that cores would jump to many different frequencies, never really settling on what was suggested as the top Turbo Core clock rate. As a result, performance gains attributable to Turbo Core seemed more modest than what I expected.

Fortunately, AMD says it made some changes to the technology for Bulldozer that should improve its effectiveness compared to Thuban.

FX Does Turbo Core A Little Differently

Application Power Management (APM) describes Zambezi/Valencia/Interlagos’ ability to monitor (in real-time) the amount of power each core consumes. Rather than taking thermal or current measurements, the activity of each Bulldozer module is tracked. AMD knows how much power each operation requires and is able to come up with instantaneous power use on a per-module basis. A quick comparison between real consumption and maximum TDP indicates whether or not there’s headroom to increase performance. In an example where you’re running an application that doesn’t tax the processor’s resources, Turbo Core dithers between the processor’s base frequency and a higher clock rate, jumping between them to average better overall performance at the defined TDP.

Turbo Core isn’t limited to just a base and some arbitrarily higher frequency, either. It’s actually implemented in three p-states: the base (referred to as P2), an intermediate state (P1), and a higher state (P0). That’s an improvement over the first-gen version of Turbo Core, which AMD says only switched between two p-states. And it’s significant, too, because you can enter P1 with all eight cores active, so long as the headroom is there. Stepping up to P0 requires at least two of four modules to idle. AMD does allow the chip’s TDP to be exceeded instantaneously, but of course it can’t hold that for any thermally significant amount of time.

As such, when you look at the specs for an FX processor and see CPU Base, CPU Turbo Core, And CPU Max. Turbo, you are guaranteed to always get at least that base frequency. You’ll see the Turbo Core clock rate so long as TDP is in check (as it would be in a well-threaded workload that doesn’t exceed the processor’s thermal ceiling). And, whenever half of the chip’s cores are idle, it’s possible to realize maximum Turbo Core speeds.

In the top chart, we see Turbo Core’s effect on iTunes, a single-threaded title. Because seven of its eight cores are essentially idle in this metric, the FX-8150 is allowed to dither at up to 4.2 GHz (it doesn’t hold that frequency constant; rather, it bounces between P1 and P0, or 3.9 and 4.2 GHz). The result is a 10-second shave compared to the same test running without Turbo Core turned on, yielding a flat 3.6 GHz.

The chart below is indicative of 7-Zip, a more thoroughly threaded application able to tax all of the -8150’s resources. Again, you don’t get a constant 3.9 GHz. With Turbo Core enabled, FX-8150 dithers between 3.9 and 3.6 GHz (versus a straight 3.6 GHz with the feature disabled). The resulting two-second speed-up is pretty modest. Even still, we have to appreciate the “free” performance that wouldn’t have been possible in first-gen Turbo Core limited to two p-states.

Ask a Category Expert

Create a new thread in the UK Article comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 16 comments.
This thread is closed for comments
  • 5 Hide
    jaksun5 , 12 October 2011 15:14
    Fuck this, I'm over your comments sections. They either don't work, are flooded by spam or I lose everything I've written and have to rewrite
  • 3 Hide
    jaksun5 , 12 October 2011 15:21
    OK, here it goes again... :-)

    Unfortuante that there wasn't a more competitive showing by AMD. Up until recently we could still say that performance pre dollar was still with them in alot of cases. Now it seems even that point is going to Intel for some time to come.

    One the bright side it appears that here in Oz that a new segment in the full size (14-15") notebook market in the last few months created by the release of the AMD Radeon on die processor powered notebooks in the $330-$450 space, where previously new notebooks could barely be had under $500, and even then they were powered by awful Celeron processors with even worse graphics. If AMD can move enough of these low end units then maybe they'll have a chance to improve their line up, if the talk of scaling isn't just hot air.
  • 4 Hide
    bobbyp86 , 12 October 2011 15:23
    Looks like I've saved myself a load of money upgrading my x4 955 this year, Bring on the 7000 series GPUs :D 
  • 3 Hide
    technogiant , 12 October 2011 15:59
    AMD is becoming a "promising pete", it's always jam tomorrow but NEVER performance delivered today.
    Be that with their roadmap of promised performance increases or the promise of increased performance on apu's via gpgpu applications.
    I will believe it when I see it if ever.

    I don't think they are even plan that effectively, I mean their proposed utilization of core/module parking in win8, great for power efficiency, but what about performance? For that you would need to spread the threads evenly across the single cores of each module so they don't share resources and only start using the second core in each module when the first core approaches max load.

    The implementation in win8 will only reduce performance and enhance power efficiency.
  • 1 Hide
    doive1231 , 12 October 2011 18:41
    I feel like the hyper-intelligent pan-dimensional beings who have just been given the answer of 42 to their question ie. disappointed and fed-up I have to wait for something better. Perhaps we should leave it to Intel to build a computer capable of finding the question.
  • 2 Hide
    blubbey , 12 October 2011 18:53
    I would say I'm disappointed but it's not like we didn't know this already - surely there'd have been some 'leaked' benchmarks on the internet to promote it more if it was as good, if not better than SB.
  • 2 Hide
    codefuapprentice , 12 October 2011 20:47
    I'm actually disappointed in bulldozer, i was hoping it would give intel a massive shake up like the athlon series did for a few years, as it stands i'm not gonna be upgrading from my Phenom II 955 any time soon
  • 2 Hide
    das_stig , 12 October 2011 22:56
    Not the best review for AMD but look on the bright side, the prices will drop like a stone and aslong as it can play all your games at the highest resolution and all the eye candy on, without needing its own power plant and pipeline to the south pole for cooling, then why worry.

    Can we all afford these super computers sucking 1000 watts from the socket, no, I would rather wait a fraction of a second and save a few quid each month.

    Future chips may just come with a few surprises, once AMD wake up and smell the coffee.
  • 1 Hide
    Anonymous , 12 October 2011 23:26
    Well common boys don't expect AMD to come out of the blue and own SB. AMD is in a very different situation, they went the GFX route awhile back and hence much of potential RnD money was taken away. Intel simply spends huge amounts of cash on their manufacturing process and micro-architecture development, which is why it's leading atm. IMSO (in my subjective opinion) Bulldozer was a strategic move intended to compete in the long run, so perhaps we will see what comes of it.
  • 1 Hide
    wild9 , 13 October 2011 00:36
    I really don't know what to say about Bulldozer, I've got very mixed feelings. In the meantime thank you Chris Angelini, for the in-depth analysis.
  • -1 Hide
    jrtolson , 13 October 2011 05:45
    to all those expecting the bulldozer to be a "holographic chip from the future, running at 200 ghz" to have turned itslef on browsed porn for u before u got home from work? then u are fools lol (no offence)

    im pretty sure the current business models for both amd and intel are not "spend 200 squillin dollars" in r&d on making processor chips that can run the main computer of a galaxy class starship, using exotic materials (other than silicon) etc etc

    im running an amd64 3200+ single core (venice) in my rig, and it does everything and more than i want it to do.. i can play all the latest games run the most demanding software.. my point is my pc 7 years old im running windows 7 on it and it does me fine.. the market does not need nor are consumers ready for a leap in processor tech so for a business model why not realease minor improved chips and keep the dollars rolling in? than gamble everything on something that might break your company before it is even ever realeased?
  • 1 Hide
    theFatHobbit , 13 October 2011 07:16
    I was hoping this would make intel nervous and lower their prices to compete with bulldozers price/performance. but no luck.
  • 0 Hide
    miklatov , 14 October 2011 02:37
    These results are a real shame. I'm neither AMD or Intel inclined, prefering to stay agnostic, but I do like healthy competition (It works well for us buyers, right? :D ) and this offering just doesn't really cut it on performance or price.


  • 0 Hide
    dillyflump , 15 October 2011 03:43
    Have to say i'm a little disappointed at the raw power per core of these FX chips in games, but i'm pretty sure the intel sandybridge and other core i7's are out in front due to hyperthreading on each core. World of Warcraft is programmed to only use two physical cores, but the intels get around it with hyper threadings 2 extra logical cores to process on. If game engines were better programmed to actually work on a cpu's physical cores and not logical ones i'm pretty sure the FX chips would beat the sandybridge processors. Perhaps the tested could look it up, but last year I was reading an article on how to force the warcraft engine to use multiple cores not just the 2. Looked complex to do but having ordered a bulldozer FX 8-core and a new 990FX board i think i'll try and get this work around to use all the chips power and see what results i get teamed with crossfire 6870's
  • 0 Hide
    HEXiT , 17 October 2011 10:35
    lolz... seriously m8 try to at least understand... the 2500K doenst support hyper threading so how can it be out in front because of it... AMD promised the world a cpu that could compete with intel's latest and they delivered 1 that can compete with there last gen only. as for you being pretty sure... well im pretty sure you think you know a lot more than you actually do, and its gonna cost you a fair bit of cash...

    not only does the part not perform consistently and never will in a gaming environment. its power inefficient to the tune of 180+watts. seriously guy rethink your choice... you would be no worse off performance wise buying a P'II 970 and waiting for the next iteration that will still underperformed against intels ivy bridge...

    as for your theory on how WOW is processed your off the mark there too.. intel only use hyper threading when a game/application asks for it. on a single core wow will use hyper threading (if available) as i needs 2 Cores to work best, on a dual core it will use 2 cores without hyper threading and on a quad it will use 2 cores without hyper threading. just because a core shows 75 percent usage doesn't mean its using 25 percent hyper threading.
    case in point wow performs no better on an intel 2500k than it does on the intel 2600k 1 has hyper threading the other doesnt.

    seriously m8 i aint trying to be a jerk, but it defiantly looks like you have a case of "thinks he knows"... you seem to be operating on assumptions about intel rather than fact... use places like wiki, toms, hardware secrets and other places to get the rite info b4 you make a misinformed choice.
  • 0 Hide
    Anonymous , 12 November 2011 22:48
    But between the i5 and i7 which is best on cost vie performance to be honest i have not look at an AMD chip based PC in years, why would you? and based on the excellent review / bench mark i will not be changing my mind for some years to come.