• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

6 or 8 core Steamroller based AMD CPU likely?

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
For what I do, Intel's HT wouldn't help me at all. I would buy an i5 over an i7 any day for this reason. IBM's POWER8 having 12 cores, which can produce up to eight threads per core with massive caches all over the place -- it's pretty crazy and offers ridiculous throughput. I don't care about die space, so yeah, more impressed by IBM in theory.
 
AtenRa said:
No its not, you need a wider execution unit and much more in the front-end in order to have any meaningful performance out of it.
Click to expand...

Nehalem wasn't really wider than Conroe at all, but the hyper threading still provided a performance boost a lot of the time. Likewise, the substantial increase in width from Yonah to Conroe was accompanied by a substantial performance benefit.

In other words, that level of execution width is utilized by one thread enough of the time to be beneficial, and underutilized enough of the time to make throwing another thread there worth it.

Incidentally, it's not really about width in an absolute sense; Bonnell/Saltwell were quite narrow but benefited tremendously from SMT.
 
Exophase said:
I assume you meant L1I, not L1D. Where are you hearing that the icache will be partitioned per core? This is the first I've heard something like that.
Click to expand...
Yeah, I meant L1I.
Bulldozer/Piledriver has a 2-way set associative 64KB instruction cache like all of its predecessors going back to the K7 (and even the K6 had 2-way set associative 32KB instruction cache). The natural extension to get 96KB would be to move to a 3-way set associative instruction cache; incidentally this is what Cortex-A57 is doing too. You can't partition this into two separate 48KB caches, you can't have 1.5-way caches.
Click to expand...
You're almost undoubtedly right. I was claiming that because the way the cache is arranged on die, at least with that leaked (supposedly) Excavator photo, it's in two separate arrays, with one nestled next to each decode stage.
 
ShintaiDK said:
Hyperthreading adds less than 5% to the core for 10-25%. Thats essentially free.
Click to expand...

Not for the consumer though. 😉 They have to pay almost 50% more for the 4770K compared to the 4670K, despite there only being a 5% die area difference. That sure is easy money for Intel.
 
Fjodor2001 said:
Not for the consumer though. 😉 They have to pay almost 50% more for the 4770K compared to the 4670K, despite there only being a 5% die area difference. That sure is easy money for Intel.
Click to expand...

The dies are the same size for a 4670K & 4770K.

HT is just turned off in the 4670K.
 
Fjodor2001 said:
Not for the consumer though. 😉 They have to pay almost 50% more for the 4770K compared to the 4670K, despite there only being a 5% die area difference. That sure is easy money for Intel.
Click to expand...

Do you understand that manufacturing cost =! Price, right?
 
^To be fair I think both AMD and Intel have been doing that for ages. Some i5's may not have properly-working caches or HT so they simply fuse them off and rebadge them as a lower product. Although I suspect that's only the case for a small fraction of the total chips.

AMD did the same with the market demand for their Phenom II chips... Remember the days of core and cache unlocking? Buy a Phenom II x2 and unlock it to a quad-core...
 
PPB said:
So you are admitting they are artificially locking HT on what otherwise would be a full blown i7?
Click to expand...

I have nothing to admit.

I am stating however it is my belief that they are the same dies, with HT fused off or whatever they do to prevent it being available to users.
 
NaroonGTX said:
^To be fair I think both AMD and Intel have been doing that for ages. Some i5's may not have properly-working caches or HT so they simply fuse them off and rebadge them as a lower product. Although I suspect that's only the case for a small fraction of the total chips.

AMD did the same with the market demand for their Phenom II chips... Remember the days of core and cache unlocking? Buy a Phenom II x2 and unlock it to a quad-core...
Click to expand...
Yep. Bulldozer/Vishera's the same way, AFAIK. I think core unlocking is still around.
 
CHADBOGA said:
I have nothing to admit.

I am stating however it is my belief that they are the same dies, with HT fused off or whatever they do to prevent it being available to users.
Click to expand...

This isn't something dramatic that you have to be ashamed to "admit." It's just a fact of how Intel does business. I guarantee you that the 4670K core and 4770K core are 100% the exact same silicon, just with some features/cache fused off.
 
Fjodor2001 said:
Not for the consumer though. 😉 They have to pay almost 50% more for the 4770K compared to the 4670K, despite there only being a 5% die area difference. That sure is easy money for Intel.
Click to expand...

Its not HT that cost you 100$, its mainly the 2MB extra cache and the speedbin. Just like the difference between 4930K and 4960X. 100Mhz, 400$. Its easy to forget for some people I can see.

But again, I was talking about diespace with production cost.
 
Last edited:
PPB said:
So you are admitting they are artificially locking HT on what otherwise would be a full blown i7?

They also fuse off 2mb of L3 cache
Click to expand...

So is all companies because everyone are not willing to pay the full price when they want a different product, yet the production cost of creating an entire new product is too high. Its how capitalism works.
 
Last edited:
Same as when you have to disable CMT with AMD because it runs worse.
Click to expand...

You shouldn't have to disable CMT (i.e. disable 1 core per module), just make sure that you have the scheduling fix for Win7 (comes with Win8 and above).

@NTMBK: Thanks for the link. Very interesting.
 
NaroonGTX said:
You shouldn't have to disable CMT (i.e. disable 1 core per module), just make sure that you have the scheduling fix for Win7 (comes with Win8 and above).

@NTMBK: Thanks for the link. Very interesting.
Click to expand...

Can you link me the results that win8 fixes the scaling issue? Because as far as I know it didnt.

ThreadsvsBenchwellat4GHz.png
 
I never said it fixed the scaling issue. I said it fixed the scheduling issue (as in the scheduler hotfix comes with Win8 already, it doesn't need to be applied.)

The scaling problem isn't something that can be fixed via software, it needs to be remedied via hardware, and that is Steamroller.
 
NaroonGTX said:
I never said it fixed the scaling issue. I said it fixed the scheduling issue (as in the scheduler hotfix comes with Win8 already, it doesn't need to be applied.)

The scaling problem isn't something that can be fixed via software, it needs to be remedied via hardware, and that is Steamroller.
Click to expand...

So...disable CMT or get a performance penalty.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top