Retail Phenom II X6 1055T "TurboCore" preview

[informal]

XS just got one retail 1055T in AMD section. User imamage got one in retail store :
http://www.xtremesystems.org/forums/...d.php?t=249434

Also there is one 1090T in AMD section as well!

[Manicdan]

power gating per core should make some incredible laptop cpus, where every watt is really noticed.

[OhNoes!]

@ Informal: Finally, you admit there's a problem with the implementation, even if while being sour and sulky at the same time! I don't know how you do it, dude.

[ryboto]

Switched off, powered down, what's the difference? The thing is the cores are not powered down (no power gating) and power could still be dissipated through leakage. Very inefficient and crude implementation at best.

Ok, so they can't power down their cores, so they lack power gating of individual cores...so what? Like I said, if the CPU can still stay in the rated TDP during turbo core boosts, what's the problem?

[ryboto]

^^ Finally, you admit there's a problem with the implementation, even if while being sour and sulky at the same time! I don't know how you do it, dude.

Why are you describing it as a problem? It's not as advanced as core gating, but it's going to work problem free and within the rated TDPs, so why are you focusing on it? Anyone following this news already knows that about the x6's, so just drop it.

[Manicdan]

@ Informal: Finally, you admit there's a problem with the implementation, even if while being sour and sulky at the same time! I don't know how you do it, dude.

ok so they add turbo to a cpu that works on >2 year old motherboards, and u still think theres problems? or how about they did it without having to start from scratch.

they just gave phenom a new feature, and are pricing it well within anyones reach, but thats just not enough?

[informal]

@ Informal: Finally, you admit there's a problem with the implementation, even if while being sour and sulky at the same time! I don't know how you do it, dude.

Oh noees!, Houston we have a problem!!! /panic mode off
First ,the "problem" is not a big issue since cores are in low clock mode,they do draw power even then but not nearly as much as you would think. Second , "problem" becomes and advantage when you do not need to have some additional circuitry inside your chip to average the temp. readings in your MPU in order to know if your chip can actually use Turbo or not. With AMD there is one htc limit temp. than is the limit for any chip they make. Turbo CORE will engage always and won't depend on the cooling solution or cramped case one is using. Even with the def. heatsink and fan it will engage if one condition is met : OS is not using half of the cores.

[crazydiamond]

I have already said that the x6 line is good, and that they don't need turbo to compete. I guess fanboys like you are just averse to any type or form of criticism from any quarter. If even people like Dresdenboy think it's a problem, then why can't I say AMD could have done better without TC implementation at all, or improved the current implementation?

could it be better, sure, everything has room for improvement, if it didnt we would never get new hardware. to say that they would have been better off with no turbo is just hating. 3.3ghz for poorly threaded apps is better than 2.8ghz no matter how you slice the bread. if they stay under tdp who cares how they implemented it. just let it go man, let people be happy about the new hardware - no reason dump all over amd because its not how YOU would have done it

edit: lol i see the last 3 posts all pretty much saying the same thing. hopefully we can move along now.

[Manicdan]

edit: lol i see the last 3 posts all pretty much saying the same thing. hopefully we can move along now.

sorry but i have this bad taste in my mouth now and nothing can make me hap... ooooo look at the pretty benchmarks

[ajaidev]

I just wanted to point out one thing.

AMD's 3 core Turbo core in 1055T pushes 3 cores 500mhz more.

Intel's Turbo Boost pushes 1 core 540mhz more in i5 750 and 660mhz more in i7 860

Finally Intel's Turbo Boost pushes 1 core just 260mhz in case of Bloomfield i7 930.

The difference in turbo is so much that a i7 870 can out perform a i7 950 in some cases. 1055T on the other hand is very different and unlike the Intel ones the dual state only process will help it greatly in situations that require two cores.

While two cores are active 1090T does 3.6Ghz, 1055T does 3.2Ghz, i7 870 does 3.4Ghz, i7 860 does 3.3Ghz, i7 930 does 3Ghz and i5 750 does 3.2Ghz

EDIT: Forgot to include the freq if 3 cores are active instead of 2. i7 870 does 3.2Ghz , i7 860 does 2.9Ghz , i5 750 does 2.8Ghz , 1055T does 3.2Ghz and 1090T does 3.6Ghz.

Thus the true power of this type of a turbo lies in dual/triple threaded application. Just imagine a 1090T running at 3.6Ghz @ 3 cores while i7 870 runs at 3.2Ghz @ 3 cores, difference between the two is 400mhz and by the looks of it if a application uses three threads it will be very beneficial for these processors.

[zalbard]

from 1.4000 to 1.475v is a big jump?

Yes, that is quite a lot.

[HousERaT]

<-------- hope these chips will give Intel a reason to rethink their pricing strategy.

[Manicdan]

Yes, that is quite a lot.

VID and voltage should not be mixed up. i dont think that a .1v increase to VID means .1v increase to core voltage. i think the VID exists so you can fine tune the voltage. but im not chew or informal, or one of the other AMD experts, so please dont assume im right.

according to Dresdenboy, its 1.225 stock, and 1.425 when Turbod. these values might be different for each model though, i expect the 95W chips to be like 1.2v and 1.35v

[ajaidev]

I did not include core 5 and 6 because it same as the 4 core for the Thuban.

[FlanK3r]

ajaidev: u compared 32-bit vs 64-bit...64-bit is more better clock for clock

[ajaidev]

^ ya i know did not see it when i posted, must be the spree of the moment

[Manicdan]

^ ya i know did not see it when i posted, must be the spree of the moment

if were talking about CB11.5, on my 940 its like 3.35 vs 3.5 (i think, i did it yesterday and im at work so i cant tell, but i think the difference is like 5%, unlike CB10 which is 25%)

[FlanK3r]

omg...now bit OT, but think Bulldozer will terror of Intel , AMD going more and more UP

[saaya]

It seems everytime I check back, something new pops up about turbo core that makes it seem rushed and poorly implemented. The biggest deal-breaker (IF TRUE; check out "Turbo CORE 101" thread) is that during turbo boost, all the idle cores are still sucking the same juice as active cores! That's dropping the ball right there! That means even in single core turbo situations, 5 idle cores would be sucking as much as 1.5v+ while doing nothing.

IF ALL THESE TURN OUT TO BE TRUE, then, IMHO AMD should not have implemented turbo in the first place. It makes them look more like amateurish copycats who would rush a badly implemented feature to market just to score some few benchmarking points. The x6 line is already a very good bargain (compared to Intel offerings) without the Turbo CORE feature! And why not just call it Turbo Boost? I mean the cores are boosted during this stage and not CORED or anything Regardless of nomenclature, I'm sure Intel still gets to collect their tax. So all in all, very tax-unfriendly, and environmentally unfriendly implementation. But I'd still get it for a dedicated encoding system, unless Intel brings an affordable hexacores to market some time soon.

well, better this than no turbo core at all imo...
but yeah, it does seem rushed and for mobile cpus and power conscious users its not a good feature... and in servers it might not be that great either, thats where turbo boost also seems to make a lot of sense as it reduces idle power consumption... but doesnt amd have at least 2 power planes on their cpus? so if half the cores or more are idle im sure they at least reduce vcore if not even shut it off?

and the idle cores will consume power and wont be turned off like on intel cpus, which means higher power consumption for the amd cpu, right? well overall they consume about the same power... intel uses the power more efficiently, but in the end what matters is overall power consumption and performance... if amd can be competitive there with a worse implementation then who cares?

[Lightman]

To everyone complaining about lack of Power Gating:
Power Gating requires special type of transistors which are not available on GloFo 45nm process. Therefore AMD couldn't design CPU on this process with Core Gating technology. This is planned for 32nm process.

One thing to remember is that AMD decided to introduce Turbo Mode only using improved 45nm process from GloFo. They could easily copy Turbo even with Rev. C3 of Phenoms but they decided to wait for Lower K dielectric which limits leakage.

I hope this clears things up a bit.

[vitchilo]

So it should be available in two weeks... I can't wait! I want benchmarks before that...

[PatRaceTin]

did amd can lock at turbo multi

like intel board?

[insurgent]

This will probably be the first AMD cpu I'll buy for my main rig after 4 years or so.

[saaya]

Yes, that is quite a lot.

its a 9% boost in voltage... hows that a lot? or do you mean 1.475v is high? on the latter i agree, but a 9% boost in voltage isnt much imo...

did amd can lock at turbo multi

like intel board?

no, intel has a turbo multiplier even for all cores active... i dont think amd does.
intel only did that cause if they would run the cpus at the 4core turbo all the time, they would be above the 125W tdp

[Manicdan]

its a 9% boost in voltage... hows that a lot? or do you mean 1.475v is high? on the latter i agree, but a 9% boost in voltage isnt much imo...

power consumption is powered (^), so 9% is really like 20% more power draw. to put it in perspective, with CnQ, if i have all 4 cores at 1.0v 800mhz maxed out load, it still consumes less power than when at 3.0ghz 1.35v idling. boosting that to 1.5v when i was water cooling, i think it drew over 100W more, from the wall, so like 65-70W more on the cpu