AMD Phenom X6 1090T Black Edition & 1055T launch on 4/27

[BulldogPO]

Any idea how these might perform on WCG?

[Manicdan]

I've seen over 350w at the wall in certain circumstances with a "minimal draw vidcard in 2d" @ 6+ on 790.......I'm sure the 890 boards will handle this fine.

whats that @ 6+ mean? the ghz?
i think that would be a special occasion since the cold air from LN2 might be good to keep the board cold too. im thinking water cooling fans might have to buy water blocks for the PMW, were talking about 40% more power draw from having 50% more cores.

im just saying we might see a thread in the AMD section listing what the limits are for boards. i already saw one set fire myself, lol

[haylui]

whats that @ 6+ mean? the ghz?
i think that would be a special occasion since the cold air from LN2 might be good to keep the board cold too. im thinking water cooling fans might have to buy water blocks for the PMW, were talking about 40% more power draw from having 50% more cores.

im just saying we might see a thread in the AMD section listing what the limits are for boards. i already saw one set fire myself, lol

I believe any boards supporting 140W TDP has no problem allow Phenom II x6 to run. The problem is that IF motherboard manufacturer going to update the BIOS to support the chip or not.

[Manicdan]

I believe any boards supporting 140W TDP has no problem allow Phenom II x6 to run. The problem is that IF motherboard manufacturer going to update the BIOS to support the chip or not.

i know that any 140W mobo can handle a 125W hexi with just a bios update, but its how much power do these things draw when OCed. if a quad hits 180W at 1.5v 4ghz, then how much do u think a 1.5v 5ghz hexi will draw? and how many boards were built to handle such cases? id bet air coolers will not see this as an issue since the power draw turns to heat which turns to temperature which turns to instability, so that leaves water cooling, and if theres bad airflow on the PWM heatsink, id be more than willing to bet alot of people are going to be seeing issues.

but it seems everyone thinks im just being paranoid or dont understand the question.

[Monkeywoman]

More details have also emerged about AMD’s Turbo Core technology. Depending on the actual model and its specifications, AMD Phenom II X6 1000T-series chips will be able to boost their clock-speeds by 400MHz or 500MHz when only half of available cores are active, e.g., microprocessors work in triple-core mode. That said, the model Phenom II X6 1090T at 3.60GHz will most likely outperform the currently available Phenom II X4 965 Black Edition in typical cases.

source; http://www.xbitlabs.com/news/cpu/dis...rocessors.html

i thought it was going to be only a single core that would be turboed, i think this is great news to hear it will be 3 cores not 1. AMD has really matured 45nm.

[informal]

source; http://www.xbitlabs.com/news/cpu/dis...rocessors.html

i thought it was going to be only a single core that would be turboed, i think this is great news to hear it will be 3 cores not 1. AMD has really matured 45nm.

AFAIK it works in four core mode too(4 active cores/threads). It's one bin over stock in that case though.

[vitchilo]

From what I've read, they switched the 1090 at 3ghz for the 1095 3.2 ghz... So the 1090 will still be out, but in Q3 instead of Q2. And the 1095 will be out at Q2... anyone can confirm?

[informal]

It's been posted and discussed before,please read the thread...

Phenom II X6 1090T : Thuban X6 9MB 125W 3.20GHz/3.60GHz AM3 Q2 2010

[dess]

From what I've read, they switched the 1090 at 3ghz for the 1095 3.2 ghz... So the 1090 will still be out, but in Q3 instead of Q2. And the 1095 will be out at Q2... anyone can confirm?

Both work at 3.2 GHz. It's not clear yet what's the difference.

[gOJDO]

Both work at 3.2 GHz. It's not clear yet what's the difference.

1095 will have a little beat more aggressive turbo mode.

[naokaji]

and if theres bad airflow on the PWM heatsink, id be more than willing to bet alot of people are going to be seeing issues.

but it seems everyone thinks im just being paranoid or dont understand the question.

I see the problem, but its easy to solve, watercool the pwms

[Monkeywoman]

1095T has been mentioned through this thread but doesn't appear on the official road map...

[informal]

Fitting 6 cores under 125W is pretty darn impressive job if we consider the 3.2/T3.6Ghz and 45nm SOI process!
Now these will surely go under LN2 sooner or later,there should be no cold bug as with the deneb cores.

[Hell Hound]

I want 4ghz 24/7 w/am2+ mobo.

[ajaidev]

I forgot why there was 125w and 95w 1055T coming out at the same time?

What was the diff again?

[informal]

I forgot why there was 125w and 95w 1055T coming out at the same time?

What was the diff again?

That should be an equivalent to standard(125W) and HE(highly efficient,95W?) Opteron models.

[Oliverda]

Fitting 6 cores under 125W is pretty darn impressive job if we consider the 3.2/T3.6Ghz and 45nm SOI process!
Now these will surely go under LN2 sooner or later,there should be no cold bug as with the deneb cores.

Exactly. Few weeks ago I said that a 3GHz/140TDP modell could be the top until Q4. So this 3.2GHz/125W at the start is a surprise. GF has made the home work. It was very important for the AMD beacuse same dies will be under the Opteron 6100 series CPUs cap.

Anyway I'm expecting at least ~6.5GHz bench stable results under LN2.

[informal]

Exactly. Few weeks ago I said that a 3GHz/140TDP modell could be the top until Q4. So this 3.2GHz/125W at the start is a surprise. GF has made the home work. It was very important for the AMD beacuse same dies will be under the Opteron 6100 series CPUs cap.

Anyway I'm expecting at least ~6.5GHz bench stable results under LN2.

Indeed,those Magny C. should do great in server arena . Thuban/Zosma with current starting power bands and clocks are giving AMD a chance to even fight in the upper perf. desktop segment (just below Gulftown). And with the aggressive pricing we have seen and socket compatibily these cores are destined to succeed!

[haylui]

i know that any 140W mobo can handle a 125W hexi with just a bios update, but its how much power do these things draw when OCed. if a quad hits 180W at 1.5v 4ghz, then how much do u think a 1.5v 5ghz hexi will draw? and how many boards were built to handle such cases? id bet air coolers will not see this as an issue since the power draw turns to heat which turns to temperature which turns to instability, so that leaves water cooling, and if theres bad airflow on the PWM heatsink, id be more than willing to bet alot of people are going to be seeing issues.

but it seems everyone thinks im just being paranoid or dont understand the question.

motherboard with less than 4+1 phase power supply circuit on board are at the lower end, thus this is not recommended for overclocking. While those overclocking friendly board would have large piece of heatsink on PWM units. So with good casing air flow I guess this is not a problem.

[Dresdenboy]

I forgot why there was 125w and 95w 1055T coming out at the same time?

What was the diff again?

30W

SCNR

Well, as informal said, very likely simply two models with different TDP at different price points.

[FlanK3r]

Hm, and 1Q 2011 Bulldozer :-D. Jupiiiii .
PS: yes, TDP 125W for 3.2 GHz hexacore, 45nm and SOI process is brliliant !

[Macadamia]

I remember people *cough* claiming that the six-cores would all run <2.6Ghz (the usual intel suspects btw)


As for the 1055T 95W vs 125W, it's probably that:

@ Base clock with full usage power load is <95W, but 125W gives more possibilities for turbo (if they are calculating by TDP/Power usage).

Pricing difference shouldn't be great, it's just that you potentially lose single threaded perf with the 95W version.

[Dresdenboy]

As for the 1055T 95W vs 125W, it's probably that:

@ Base clock with full usage power load is <95W, but 125W gives more possibilities for turbo (if they are calculating by TDP/Power usage).

Pricing difference shouldn't be great, it's just that you potentially lose single threaded perf with the 95W version.

But the power consumption in stock clock and turbo clock modes (P-0 and P-Max) is strongly related (as I've calculated on my blog lately). So if the 1055T is supposed to run 6 cores at 2.8GHz at <95W, then it will do that also with 3 cores at 3.3GHz and the others idle.

[informal]

I remember people *cough* claiming that the six-cores would all run <2.6Ghz (the usual intel suspects btw)


As for the 1055T 95W vs 125W, it's probably that:

@ Base clock with full usage power load is <95W, but 125W gives more possibilities for turbo (if they are calculating by TDP/Power usage).

Pricing difference shouldn't be great, it's just that you potentially lose single threaded perf with the 95W version.

We all know them ,we just don't mention their names now(it would be even more embarrassing ).

125W vs 95W looks like binning job to me

[Manicdan]

think there will be a nice way to modify the pstates so we can set our own turbos? im a fan of overclocking, but also power saving, and its been a real pain to keep CnQ on while OCing.