[WCCF] CPU-Z shows AMD Carrizo APU A10-8890K

[LukaP]

+1

 

whole buldozer series were like that. fake cores!  they just couldn't make technology like intel HT and made modules with 2 real cores that work like one

no. 

 

they had the tech, and the people. (if you look at the phenoms before, those had true cores)

 

in essence there are two ways one can go about multithreading. one is SMT which is what intel went with, which provides superior IPC but is harder to have a lot of cores, or CMT which is what AMD uses now, which doesnt to singlethreaded well, but is really easy to scale to more and more modules/cores.

 

CMT  SMT

[Linus_torvalds]

no. 

 

they had the tech, and the people. (if you look at the phenoms before, those had true cores)

What no? You read my post and understand nothing? I wrote about buldozers, phenomes are not buldozers. each buldozer CPU has modules with 2 cores and thats very bad.

[LukaP]

What no? You read my post and nothing understand? I wrote about buldozers, phenomes are not buldozers. each buldozer CPU has modules with 2 cores and thats very bad.

no to "they couldnt make technology like intel HT"

 

they could. all they had to do was add another registry set to the core, increase the cache and rewrite some microcode to make HT. but they chose to go a different route to see what happens. sadly they didnt do so well

[HerEscutcheon]

Are you guys not reading my post above?  :huh:

Nobody actually reads anymore, you should know that!

[Admiaczking]

oh i didn't notice the WCCF, nevermind

i'm gonna go sob in a corner while i wait for better FX processors

[AnnoyedShelf]

Daaaamn. 20nm. Nice work AMD.  EDIT: Nevermind.

[Admiaczking]

Daaaamn. 20nm. Nice work AMD.  EDIT: Nevermind.

way for WCCF to be the biggest killjoy of all time eh?

[AnnoyedShelf]

way for WCCF to be the biggest killjoy of all time eh?

Nothing that comes out of WCCF is legit anymore.

[Admiaczking]

Nothing that comes out of WCCF is legit anymore.

i doubt any of it ever was

[Linus_torvalds]

no to "they couldnt make technology like intel HT"

 

they could. all they had to do was add another registry set to the core, increase the cache and rewrite some microcode to make HT. but they chose to go a different route to see what happens. sadly they didnt do so well

you are not right

 

they could not. because if they did then their core wich is 2x weaker then intel's core would have become weaker

[LukaP]

you are not right

 

they could not. because if they did then their core wich is 2x weaker then intel's core would have become weaker

im not talking about putting HT into their bulldozer type cores... im talking about making their SMT oriented cores that were in the phenoms be able to process more threads per core. they couldve done that, and it would 2x boost those cores in multithreaded tasks, but they chose a different more risky route and it didnt work out

[Linus_torvalds]

and it would 2x boost those cores in multithreaded tasks, but they chose a different more risky route and it didnt work out

you do not understand HT. it's not 2x boost. core does some task and there is some horse powers left so with help of HT it does that second task but not so fast as first one. 

[ACatWithThumbs]

Can we just stop posting stuff from WCCF on this forum.
They never have any legit sources.

 

[PortentousLad]

well, AMD finally made some good APUs! i might actually... BUY ONE!

<Highly subjective opinion do read with caution.>

Boloks the 7850k is awesome do not regret buying it one bit.

But seriously it's been doing a fantastic job.

Oh god! I knew it would get in here eventually. People- this is a fake. A few guys on Overclock.net did this to see if "journalists" would do the research before publishing it: http://www.overclock.net/a/a-troubling-trend

Guru3D was the first to bite, and they actually plagiarized as well. WCCF then essentially copied it and added some of their stuff to it on their main website and a sister rumour mill. Soon other places put it up also. The only news here is that the tech journalist world is full of people who would rather post first for views and exclusivity than take the time to fact check.

That's hilarious!

[Silent_Xaxal]

Am so hoping that they have DDR4 on their APU platform/s soon. Intel doesn't have a cheaper DDR4 platform for the next year or so. Then again, DDR4 will come out of the gates as pretty expensive.

[LukaP]

you do not understand HT. it's not 2x boost. core does some task and there is some horse powers left so with help of HT it does that second task but not so fast as first one. 

i know its not a 2x boost... its a theoretical 2x boost if correct code is written. and the second task, when its being executed is executed at the same speed, its just that its only executed when the first task is waiting...

[samcool55]

It's fake, this is the 1 million post on overclock.net so don't worry about this

 

http://www.overclock.net/t/1504354/guru3d-20nm-amd-carrizo-apu-spotted-features-fm3/20#post_22626294

[dragoon20005]

first dead give away

 

FM3 socket... yeah right

 

next gen AMD will stay on the FM2+ socket, it in the bloody APU roadmap

 

When did APU come in 6 core???

 

 

AMD naming for the APU is in even numbers or end with 50 so should be 8800k or 8850k

[Linus_torvalds]

i know its not a 2x boost... its a theoretical 2x boost if correct code is written. and the second task, when its being executed is executed at the same speed, its just that its only executed when the first task is waiting...

I said you do not know HT well.

firstly it's not even theoritical 2x boost. 

secondly second task can be executed exectly when first task is executed. they work independently from each other on one core. That's main thing about HT. It makes core work simultaniously on 2 tasks. no one is waiting. 

You described what AMD did in their buldozer modules. each module has 2 cores, but when one executes task second is waiting. they have one source. It's like 2 line way and bridge,  wich only one car can pass through on 

[LukaP]

I said you do not know HT well.

firstly it's not even theoritical 2x boost. 

secondly second task can be executed exectly when first task is executed. they work independently from each other on one core. That's main thing about HT. It makes core work simultaniously on 2 tasks. no one is waiting. 

You described what AMD did in their buldozer modules. each module has 2 cores, but when one executes task second is waiting. they have one source. It's like 2 line way and bridge,  wich only one car can pass through on 

it is a theoretical 2x boost when your app is written to load the core when the other thread is waiting for information

 

yes they are independent. by waiting i meant waiting as in, it requested something from RAM, so its currently idle.

[dalekphalm]

I said you do not know HT well.

firstly it's not even theoritical 2x boost. 

secondly second task can be executed exectly when first task is executed. they work independently from each other on one core. That's main thing about HT. It makes core work simultaniously on 2 tasks. no one is waiting. 

You described what AMD did in their buldozer modules. each module has 2 cores, but when one executes task second is waiting. they have one source. It's like 2 line way and bridge,  wich only one car can pass through on 

Just to clear up a little bit here, the reason behind AMD's switch to the "Module" design was a very smart one indeed - they're just a little ahead of the technology.

 

Let's take the Bulldozer module and look at it:

It contains 2 integer cores and one floating point core.

 

Some people take this (like yourself), as meaning it's "fake" dual core. However, that's not entirely correct. The technology is different enough that you can't compare it directly with an Intel "Core".

 

Now, why are there half the amount of FP cores? Because many computer applications don't use them! Secondly, Graphics Cores are MUCH MUCH MUCH MUCH (did I say much?) better at performing FP calculations. This is why you have things like OpenCL and CUDA to take advantage of GPU Compute.

 

Enter the AMD APU: Combined Graphics processor and Bulldozer module

 

The idea behind this is to leverage the advanced FP power of the GCN cores inside the APU, alleviating the strain. This is the whole idea behind HuMA. The CPU portion and GPU portion can co-operate on a hardware level to perform calculations together.

 

In theory, it means it can leverage the GPU power to make the whole APU much more powerful then a traditional Phenom style CPU Core. However, the programming side of things is lagging behind significantly. Not much can actually utilize this power.

 

Now as @LukaP has mentioned, AMD took this route on purpose. They could have VERY EASILY went the HT route (Obviously they'd have to call it something else, but technologically they could easily do it). But AMD is looking at the long game, which unfortunately means in the interim, their CPU "Modules" lag behind quite a bit.

 

Also, HT does NOT do calculations simultaneously. That would be a factual lie. HT works on the principle that pretty much any calculation a CPU Core does won't fully utilize that core. It also works on the principle that once a Core has done an instruction, it is often waiting for *something* else to complete. Be that accessing the RAM, getting a result from another Core, accessing a different piece of system Hardware, etc. During these little tiny gaps, which by themselves are very small - but extremely frequent - the CPU can leverage all those combined little bits and equal a large performance boost by using the power as the Core is idle waiting for another action.

 

Hyperthreading does NOT automagically make a single core able to perform two tasks out of thin air. That's not how it works. You obviously could use some additional reading on the subject, so I suggest that you check out the wiki on HT:

http://en.wikipedia.org/wiki/Hyper-threading

 

You'll note that it says:

 

These shared resources allow the two logical processors to work with each other more efficiently, and lets one borrow resources from the other when one is stalled. A processor stalls when it is waiting for data it has sent for so it can finish processing the present thread.

 

With this in mind, this is a very very simplified explanation, and there are many technical bits missing from it.

[LukaP]

-snip-

thank you for this, i was contemplating either comitting the time to explain it properly, or just giving up

[dalekphalm]

thank you for this, i was contemplating either comitting the time to explain it properly, or just giving up

No problem. It just bugs me when people are stating factual errors. Whether it's simply ignorance of the topic (Which happens to us all occasionally, we just must try to learn from it) or deliberate trolling, either way, I was compelled to comment

[sof006]

F**k. I spilled the salt, time to do a lot of pinching because WCCF of course.

[LukaP]

No problem. It just bugs me when people are stating factual errors. Whether it's simply ignorance of the topic (Which happens to us all occasionally, we just must try to learn from it) or deliberate trolling, either way, I was compelled to comment

yeah indeed. maybe i was kindoff wrong stating some stuff, or should atleast explain why it is so.. but mostly, i cant stand seeing people here spread misinformation so much. sometimes i ignore it, sometimes i try to fix their POV, mostly i just fail, because i never have enough time all together to write up a good post on it