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

[FlanK3r]

now i know, what is x6 1095T. Think, this point is at turbo schema..
1035T +500Mhz
1055T +500Mhz
1075T +500MHz
1090T +400MHz

think, 1095T will with 500MHz with turbo

[Manicdan]

now i know, what is x6 1095T. Think, this point is at turbo schema..
1035T +500Mhz
1055T +500Mhz
1075T +500MHz
1090T +400MHz

think, 1095T will with 500MHz with turbo

i support this theory, sounds logical to me

[OhNoes!]

That's ridiculous. You mean AMD is actually going to rate processors based on the "gimmick" called Core (Turbo) Boosting? :p

[Lightman]

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

It also looks like SFF or HTPC sex-core CPU

@FlanK3r

Just look at my post several pages ago
http://www.xtremesystems.org/forums/...8&postcount=50

Originally Posted by ajaidev View Post
I dont understand the 1090T codename??

1035-1055-1075 and then 1090 ?? why shouldn't it have been 1095T

I think is has something to do with 'only' 400MHz turbo compared to 500MHz for 1035-1055-1075 models maybe?
Just a guess ...

[gOJDO]

@Ohnoes!
For what gimmick are you blabbering? That Core (Turbo) Boosting will boost performance of their processors (running at specs). Thanks to that feature Nehalem is so better than Phenom II and Core2 Quad at "same" clock.

[FlanK3r]

That's ridiculous. You mean AMD is actually going to rate processors based on the "gimmick" called Core (Turbo) Boosting? :p

yes, and as u watch, im not alone (im not noob at CPU AMD man -some years ago interesting for it )

Lightman
Heh, thanks, now we are 2 mans here with the same speculation

[OhNoes!]

Since the irony is lost on you guys, allow me to clarify. It's quite funny how some condemned Turbo Boost as an Intel "gimmick" designed to win benchmarks, but when AMD implemented it, everybody is drooling all over it and are actually going to cough up for cash to settle for the higher turbo on the 1095T. This is all assuming that the ONLY difference between the 1090T and the 1095T is the 100mhz extra boost when the feature kicks, since they're both supposed to be clocked at 3.2GHZ iirc?

I like the price of these processors a lot, but if the above is true, then AMD is setting a new precedent here (cough, cough) - I don't want Intel getting any ideas :p.

[Manicdan]

Since the irony is lost on you guys, allow me to clarify. It's quite funny how some condemned Turbo Boost as an Intel "gimmick" designed to win benchmarks, but when AMD implemented it, everybody is drooling all over it and are actually going to cough up for cash to settle for the higher turbo on the 1095T.

ill explain. a gimmick is something used to look good on benchmarks without proper representation of actual frequency and is misleading. we are happy that amd is pushing cores when some are idling, which has NOTHING do to with benchmarks. if its going to be 10$ more, then it might sell. but also keep in mind that people will probably not have a clue thats the case since sites like newegg or microcenter, probably dont list the turbo speeds, just stock speeds and model numbers.

and like u just said, were guessing, its fun to do and its not going to cause any harm, lol

[gOJDO]

OhNoes! When I red your post I thought you were serious. But now, i saw the smiley :p at the end of the post, which should have been .

I agree with you on that point. AMD fanboys will BS about every Intel feature which is not present in AMD CPUs. But once AMD implements it, they will start talking about it like it's the best invention ever implemented.

[Karolis]

now i know, what is x6 1095T. Think, this point is at turbo schema..
1035T +500Mhz
1055T +500Mhz
1075T +500MHz
1090T +400MHz

think, 1095T will with 500MHz with turbo

Hey, I have a question (sorry if it is a stupid one): from where do you know that a '1095T' is coming? I'm asking because there is no '1095T' in the leaked AMD roadmap. Anyway, I'm really impressed with the 1090T specs: 6 cores at 3.2 GHz @125W.

[msgclb]

Hey, I have a question (sorry if it is a stupid one): from where do you know that a '1095T' is coming? I'm asking because there is no '1095T' in the leaked AMD roadmap. Anyway, I'm really impressed with the 1090T specs: 6 cores at 3.2 GHz @125W.

It was first mention here.

Source

AMD will launch several six-core Phenom II processors in the second quarter including X6 1035T (2.6GHz/95W), X6 1055T (2.8GHz/95W) and X6 1075T (3.0GHz/125W), according to the sources. In the third quarter, AMD will add a few more six-core processors including X6 1095T.

[dess]

Hey, I have a question (sorry if it is a stupid one): from where do you know that a '1095T' is coming? I'm asking because there is no '1095T' in the leaked AMD roadmap. Anyway, I'm really impressed with the 1090T specs: 6 cores at 3.2 GHz @125W.

I wanted to ask the same...

(edit)

It was first mention here.

Source

I would think the 'source' mistook 1095T for 1090T, or AMD changed it's mind in the meantime (I mean, pehaps 3.7 GHz were too much of a hassle already).

I agree with you on that point. AMD fanboys will BS about every Intel feature which is not present in AMD CPUs. But once AMD implements it, they will start talking about it like it's the best invention ever implemented.

Some AMD fanboys, maybe. Just like some Intel fanboys, about the IMC, then x64 (originally named AMD64)... Anyway, I, for one, as an AMD user since a decade, never thought the Intel Turbo Boost was a bad idea. Indeed, that - for your information - the Barcelona/Agena already has the ability to clock the cores differently, and there was a little application (AMD Overdrive) with which one could set the clocks as liked. There is perhaps even a utility that can do it automatically. It's just not done in HW, as with the new CPU's.

[OhNoes!]

I wanted to ask the same...

(edit)

I would think the 'source' mistook 1095T for 1090T, or AMD changed it's mind in the meantime (I mean, pehaps 3.7 GHz were too much of a hassle already).


Some AMD fanboys, maybe. Just like some Intel fanboys, about the IMC, then x64 (originally named AMD64)... Anyway, I, for one, as an AMD user since a decade, never thought the Intel Turbo Boost was a bad idea. Indeed, that - for your information - the Barcelona/Agena already has the ability to clock the cores differently, and there was a little application (AMD Overdrive) with which one could set the clocks as liked. There is perhaps even a utility that can do it automatically. It's just not done in HW, as with the new CPU's.

AMD did not invent both of those technologies either, though. AMD was the first to bring those technologies to the desktop, however.

Having said that, I like where AMD is going with their market/price targeting. It should be reflected in their market share in the next 12 months.

[dess]

AMD did not invent both of those technologies either, though. AMD was the first to bring those technologies to the desktop, however.

I know, regarding the IMC.

64 bit, as a way to handle values, is not their idea/invention, of course. But, x64/AMD64, as a specific and substantial extension of the x86 architecture and ISA, is AMD's own invention.

[haylui]

OhNoes! When I red your post I thought you were serious. But now, i saw the smiley :p at the end of the post, which should have been .

I agree with you on that point. AMD fanboys will BS about every Intel feature which is not present in AMD CPUs. But once AMD implements it, they will start talking about it like it's the best invention ever implemented.

oh noes....I am afraid that AMD fans will BS about IMC and the ditch of FSB since K8 era and native quad core design as Intel has follow suit.

[gOJDO]

Indeed, that - for your information - the Barcelona/Agena already has the ability to clock the cores differently, and there was a little application (AMD Overdrive) with which one could set the clocks as liked. There is perhaps even a utility that can do it automatically. It's just not done in HW, as with the new CPU's.

I'm pretty aware of AMD Overdrive options. But the point is that AMD Overdrive is not for average Joe and does not come installed with the OS by default. Also, using AMD overdrive for OC-ing, practically voids your warranty. Unlike Phenom and Phenom II C2/C3, Thuban will have that feature managed by the CPU and it won't void warranty.

BTW, AMD Overdrive and "Turbo" "Boost" are for noobz and are useless for me:

[BeepBeep2]

OhNoes! When I red your post I thought you were serious. But now, i saw the smiley :p at the end of the post, which should have been .

I agree with you on that point. AMD fanboys will BS about every Intel feature which is not present in AMD CPUs. But once AMD implements it, they will start talking about it like it's the best invention ever implemented.

Please read:

http://www.gamers-crib.com/forum/sho...1&postcount=15

I hate turbo. It's retarded and it's just made for people who can't overclock.

If AMD can't run all 6 cores at a given speed it's a flop. I don't care how anyone explains it. This "turbo boost" tech from both companies is stupid. The average user will not even notice the speed increase, and whoever is doing something that will take advantage of it AND NOTICE will probably be technically inclined to the point in which they already know how to overclock the damn thing.

3.2 Ghz 125w Black Edition 6 core for $250 without turbo would be good.

[Manicdan]

Please read:

http://www.gamers-crib.com/forum/sho...1&postcount=15

I hate turbo. It's retarded and it's just made for people who can't overclock.

If AMD can't run all 6 cores at a given speed it's a flop. I don't care how anyone explains it. This "turbo boost" tech from both companies is stupid. The average user will not even notice the speed increase, and whoever is doing something that will take advantage of it AND NOTICE will probably be technically inclined to the point in which they already know how to overclock the damn thing.

3.2 Ghz 125w Black Edition 6 core for $250 without turbo would be good.

overclockers dont need it, but the rest of the world would love a 20% increase without having to do anything

[BeepBeep2]

overclockers dont need it, but the rest of the world would love a 20% increase without having to do anything

But in multi-threaded apps it doesn't do anything.

What kind of normal user needs a 20% increase in a single threaded app? Encoding is usually multi-threaded, and gaming will probably stress the CPU to where turbo doesn't even "engage" on the single core...

[JumpingJack]

But in multi-threaded apps it doesn't do anything.

What kind of normal user needs a 20% increase in a single threaded app? Encoding is usually multi-threaded, and gaming will probably stress the CPU to where turbo doesn't even "engage" on the single core...

Actually, yes it does.... +1 bin for 3 or 4 cores engaged, +2 for 1-2 cores engaged (with respect to Intel cores). The Lynnfield offer even more in the single and dual threaded scenarios. AMD's first implementation will run nominal with all cores engaged, and bump with 1-3 cores engaged. Either solution is absolutely a step in the right direction.

The move to multicore is simply a different way to take advantage of Moore's law, the physics, however, limits the clock speed not by Fmax of the design/process, but the practical limits of power and cooling. To fit a quad core, for example, within a reasonable thermal envelop, the processor is volted and clocked lower than what would otherwise be capable.

This is fine, since performance is extracted via thread level parallelism, in fact efficiency can dramatically increase if the task can be well threaded. However, Amdahl's law kicks in or some algorithms just cannot be made parallel at the thread level, as such these situations suffer with respect to the potential because any one core is clocked for the thermals.

Essentially, which would you rather have for a single/dual threaded app, a 3.4 GHz dual core or a 3.0 GHz quad core? Naturally, the best situation would be the higher clock.

So designers face a challenge, how to bring more and more cores to the masses but still yield performance on legacy single threaded applications or provide support until software can catch up.... the very natural thought is, if all but one or two cores are idle, why not just kick up the clock speed of the active cores... hence turbo modes.

It is a great idea, works as it should and gives you the best of both worlds, great performance in highly threaded environments, and a boost in performance over what could be had in lightly threaded environments.

[Hornet331]

Actually, yes it does.... +1 bin for 3 or 4 cores engaged, +2 for 1-2 cores engaged (with respect to Intel cores). The Lynnfield offer even more in the single and dual threaded scenarios. AMD's first implementation will run nominal with all cores engaged, and bump with 1-3 cores engaged. Either solution is absolutely a step in the right direction.

The move to multicore is simply a different way to take advantage of Moore's law, the physics, however, limits the clock speed not by Fmax of the design/process, but the practical limits of power and cooling. To fit a quad core, for example, within a reasonable thermal envelop, the processor is volted and clocked lower than what would otherwise be capable.

This is fine, since performance is extracted via thread level parallelism, in fact efficiency can dramatically increase if the task can be well threaded. However, Amdahl's law kicks in or some algorithms just cannot be made parallel at the thread level, as such these situations suffer with respect to the potential because any one core is clocked for the thermals.

Essentially, which would you rather have for a single/dual threaded app, a 3.4 GHz dual core or a 3.0 GHz quad core? Naturally, the best situation would be the higher clock.

So designers face a challenge, how to bring more and more cores to the masses but still yield performance on legacy single threaded applications or provide support until software can catch up.... the very natural thought is, if all but one or two cores are idle, why not just kick up the clock speed of the active cores... hence turbo modes.

It is a great idea, works as it should and gives you the best of both worlds, great performance in highly threaded environments, and a boost in performance over what could be had in lightly threaded environments.

If only the windows task scheduler was smart enough to use that potential...

[JumpingJack]

If only the windows task scheduler was smart enough to use that potential...

Ahhhhhh, I am ahead of you there ..... I have looked into this some, actually ...

Here is the upshot...

a) Vista -- dump it.
b) Win 7 -- get it.

It is very clear that Win 7 scheduler is doing a much better job, not perfect, but much better. Haven't looked at XP.

[Frontl1ne]

Say you buy a 1090T, that means you're pretty much guaranteed a 3.6GHz overclock if you know what you're doing right ?

[JumpingJack]

Say you buy a 1090T, that means you're pretty much guaranteed a 3.6GHz overclock if you know what you're doing right ?

Yeah, without question. I am particularly holding back an AM3 board for this processor, I have a X4 965 (140W unfortunately) and wall at 3.8 Ghz in a 64 bit OS, but around 4.1 GHz in a 32 bit OS. I haven't played with the newer stepping to see if this wall still exists (and I have really paid much attention to the forums to see what people are getting), but given that my (and everyone else's) X4 965 can reach 3.8 easily (at a minimum), then it is a good assumption to expect that the X6's are not Fmax bound at the very minimum, up to that point.

Jack

[dess]

But in multi-threaded apps it doesn't do anything.

According to some slides, it elevates half of the cores' clocks (while putting the rest to sleep), so 3 of them, in case of Thuban. I don't know about an application that runs only 2 or 3 heavyweight threads, and never more, though.

What kind of normal user needs a 20% increase in a single threaded app? Encoding is usually multi-threaded, and gaming will probably stress the CPU to where turbo doesn't even "engage" on the single core...

20% is not much for an overclocker, but for the rest of people, 20% increase in the clock frequency has its costs in money, until now (considering single or "lightly" threaded apps). And in CPU-bound situations in games, it can bring even that much of increase in fps, which is "a lot"...

They can't clock all the cores high, because they have to take care of the power-consumption, too - which is definitely a thing to consider, for many people. Otherways they would have to sell these CPU's with a TDP of one or two classes higher (125W -> 140W, 95W -> 125W).

Or, they could leave the clocks in the lows untouched, but then they could only sell them for rendering, videoediting, etc. and to overclockers.