Is Ivy Bridge the answer to AMD Bulldozer ?

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Phynaz

Lifer
Mar 13, 2006
10,140
819
126
Or put another way, a lower core count isnt't always an advantage over higher core count. The only reason one would argue against more cores is if one doesn't have more cores. And the point is completely moot if 8 cores do in fact out perform 4 cores, no? They are different architectures and different design choices just like Radeon and GeForce.

I know I'm going to regret replying to a zoner, but...
In what case are more cores better if performance is the the same or worse than fewer cores?
 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
I know I'm going to regret replying to a zoner, but...
In what case are more cores better if performance is the the same or worse than fewer cores?

I know i'm going to regret replying to an inteller, but....
In what case are fewer cores better if performance is the same or worse than more cores?
 

bandgit

Member
Mar 7, 2011
36
0
66
Nothing can be concluded until we see indy BD benchies, but what interests me personally is "pushing the envelope" CPUs. Cutting edge microprocessors that achieve new and previously unparalleled pure performance standards. In other words, what makes my software run faster and better than any other consumer CPU available anywhere. If BD is a valid SB LGA2011 competitor which happens to beat it to market by a few months, my credit card is burning a hole in my pocket and ready to send my hard earned money AMD-way. If BD is some "midrange" which fills in some blanks in the AMD lineup and competes with SB LGA1155, I'm BOOOOOOOOOOOOOOOOOOOORED!
 

iSeeker

Junior Member
Mar 14, 2011
6
0
0
The speed of an application is determined by the speed of the slowest thread that the app uses, and further breaking down processes into multiple threads is not always possible/feasable/optimal - it also increases 'management' costs exponential.
So single thread/core performance should remain most important in the design of CPU's.
I don't understand cpu architectures, but compromising single thread performance in favor of more cores in the name of 'modularity', is not wise seen from a programmer.

Going from 32 nm to 22 nm technology (Ivy Bridge) is another step in potentially increasing single thread performance: less power consumption can allow higher single core speeds.
Just imagine that 22nm i.s.o. 32nm technology would allow me to increase core speed by 20%.
 

Castiel

Golden Member
Dec 31, 2010
1,772
1
0
Nothing can be concluded until we see indy BD benchies, but what interests me personally is "pushing the envelope" CPUs. Cutting edge microprocessors that achieve new and previously unparalleled pure performance standards. In other words, what makes my software run faster and better than any other consumer CPU available anywhere. If BD is a valid SB LGA2011 competitor which happens to beat it to market by a few months, my credit card is burning a hole in my pocket and ready to send my hard earned money AMD-way. If BD is some "midrange" which fills in some blanks in the AMD lineup and competes with SB LGA1155, I'm BOOOOOOOOOOOOOOOOOOOORED!

Expect the latter. The 8 core bd is positioned to compete with the 2600K. As I've said numberous times. With its 130w TDP it wont even come close to sandy bridge in the overall overclocking department
 

bandgit

Member
Mar 7, 2011
36
0
66
Expect the latter. The 8 core bd is positioned to compete with the 2600K. As I've said numberous times. With its 130w TDP it wont even come close to sandy bridge in the overall overclocking department

If that is the case, then I'm not just bored, I'm catatonic. :thumbsdown:
 

Idontcare

Elite Member
Oct 10, 1999
21,118
59
91
The speed of an application is determined by the speed of the slowest thread that the app uses, and further breaking down processes into multiple threads is not always possible/feasable/optimal - it also increases 'management' costs exponential.
So single thread/core performance should remain most important in the design of CPU's.
I don't understand cpu architectures, but compromising single thread performance in favor of more cores in the name of 'modularity', is not wise seen from a programmer.

Going from 32 nm to 22 nm technology (Ivy Bridge) is another step in potentially increasing single thread performance: less power consumption can allow higher single core speeds.
Just imagine that 22nm i.s.o. 32nm technology would allow me to increase core speed by 20%.

Welcome to the forums iSeeker!

And yes, absolutely agree with you, single-thread performance improvements is the tide that lifts all boats, be they single-threaded apps or multi-threaded apps.

There is a good reason why Niagara did not take any market (be it server, big iron, workstation, etc) by storm, same reason why bulldozer won't be as simple as 1,000 486-cores bundled together.

The sweet spot for parallelization is in the range of 4-8 threads. You can capture a moderately good portion of the thread-speedup curve by operating in this regime across a wide range of Amdahl limits. Its when you start walking the thread count into double digits that you hit that point of diminishing returns for all but the most trivial of parallelizable workloads.

Single-threaded architectures may well be bound by Pollack's Rule, but multi-threaded architectures are bound by Amdahl's Law with further implications of possible performance slowdowns as you add more threads as detailed by Almasi and Gottlieb.
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
Expect the latter. The 8 core bd is positioned to compete with the 2600K. As I've said numberous times. With its 130w TDP it wont even come close to sandy bridge in the overall overclocking department

I expect the 8 core BD to beat the 2600K. Not by much, but still beat it. I also think it will be priced a little higher too. And that will be the case until s2011 comes out.

I also expect the 8 core BD to beat the 4 core s2011 CPUs. But once we get into the 6 and 8 core Intels, that is where Intel will retain the crown so to speak. But expect the prices to reflect that.

Competition is good. But the performance crown will always demand a kings ransom regardless of which company it is.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
I expect the 8 core BD to beat the 2600K. Not by much, but still beat it. I also think it will be priced a little higher too. And that will be the case until s2011 comes out.

I also expect the 8 core BD to beat the 4 core s2011 CPUs. But once we get into the 6 and 8 core Intels, that is where Intel will retain the crown so to speak. But expect the prices to reflect that.

Competition is good. But the performance crown will always demand a kings ransom regardless of which company it is.

Unless AMD pulls a rabbit out of their hat, it will ve very hard for BD to beat SB in non highly-threaded applications. Of course time will tell, but I don't see BD dominating single-threaded apps. If BD cannot dominate heavily multi-threaded applications then it really failed IMHO. That is where it really should shine with it's design.
 

Castiel

Golden Member
Dec 31, 2010
1,772
1
0
I expect the 8 core BD to beat the 2600K. Not by much, but still beat it. I also think it will be priced a little higher too. And that will be the case until s2011 comes out.

I also expect the 8 core BD to beat the 4 core s2011 CPUs. But once we get into the 6 and 8 core Intels, that is where Intel will retain the crown so to speak. But expect the prices to reflect that.

Competition is good. But the performance crown will always demand a kings ransom regardless of which company it is.

I don't expect any 4 core 2011 chips to be produced. Only 6 and 8
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
I don't expect any 4 core 2011 chips to be produced. Only 6 and 8

Anand has already said 4 core s2011 cpus will be released along with 6 core at first. (8 will follow later perhaps or wait for 22nm). And I have a reliable source also confirm this.

There is a market for them. A 4 core s2011 cpu will clock much higher than a 6 core version. And if they are priced right, say $380, then they will sell very well. It would pretty much be a 2600K with 40 lanes of PCIe 3.0, quad channel IMC, larger L3, and no IGP. I would get one for certain.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
59
91
What's the point tho idc? You can already get quad xeons for 1155
LOL, I have NFC on that, personally I don't get the dual-core xeon chip either, but presumably a market exists for it. I'm just saying I see an analogous extension of that market space, for whatever reason it exists, existing for quad-core LGA2011 sku's.

Look at itanium, intended to scale to however many sockets, and yet they sell cut-down core-count SKU's there too.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Welcome to the forums iSeeker!

And yes, absolutely agree with you, single-thread performance improvements is the tide that lifts all boats, be they single-threaded apps or multi-threaded apps.

There is a good reason why Niagara did not take any market (be it server, big iron, workstation, etc) by storm, same reason why bulldozer won't be as simple as 1,000 486-cores bundled together.

The sweet spot for parallelization is in the range of 4-8 threads. You can capture a moderately good portion of the thread-speedup curve by operating in this regime across a wide range of Amdahl limits. Its when you start walking the thread count into double digits that you hit that point of diminishing returns for all but the most trivial of parallelizable workloads.

Single-threaded architectures may well be bound by Pollack's Rule, but multi-threaded architectures are bound by Amdahl's Law with further implications of possible performance slowdowns as you add more threads as detailed by Almasi and Gottlieb.

I will agree if we only talking about one application running at any given time, but what if we run two apps simultaneously? 16 cores will be faster than 8 and what if we run 3 or four apps simultaneously etc.

AMD and Intel don’t make CPUs for Desktop/Laptop use only
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Or put another way, a lower core count isnt't always an advantage over higher core count. The only reason one would argue against more cores is if one doesn't have more cores. And the point is completely moot if 8 cores do in fact out perform 4 cores, no? They are different architectures and different design choices just like Radeon and GeForce.
That's not true and it's not what I'm saying.

If Bulldozer offers 8 weak cores, for most people 4 powerful Sandy Bridge cores would be better.

I'm just saying that if it takes 8 Bulldozer cores to match 4 Sandy Bridge cores, it makes sense to go with Sandy Bridge because then you're not relying on highly threaded software to exploit your CPU to its full potential. A large portion of modern games don't even exploit 4 cores to their potential, let alone 8.
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
I know i'm going to regret replying to an inteller, but....
In what case are fewer cores better if performance is the same or worse than more cores?

I'm not sure you understand the difference between multi threaded and single threaded applications.

Single threaded applications can only take advantage of a single CPU core. These programs usually include games, office apps, and some internet browsers.

Multi threaded applications can exploit several cores, but not an unlimited number of cores (in most cases). Some of these programs can only use two cores, for example.

So, for a large portion of the software that's out there, the fastest single threaded CPU is going to give the best performance. A good example of this is how a dual core i3 is faster than a six core Thuban in Starcraft 2.

That's why with AMD just throwing more and more cores onto their chips, it's not benefiting all scenarios equally. There are tons of applications that need better single threaded performance.

All things being equal, you'd be better off with a 4 core CPU than an 8 core CPU provided that they had the same overall thoroughput (i.e. a 4ghz 4 core CPU vs a 2ghz 8 core CPU of identical architecture).
 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
I'm not sure you understand the difference between multi threaded and single threaded applications.

Single threaded applications can only take advantage of a single CPU core. These programs usually include games, office apps, and some internet browsers.

Multi threaded applications can exploit several cores, but not an unlimited number of cores (in most cases). Some of these programs can only use two cores, for example.

So, for a large portion of the software that's out there, the fastest single threaded CPU is going to give the best performance. A good example of this is how a dual core i3 is faster than a six core Thuban in Starcraft 2.

That's why with AMD just throwing more and more cores onto their chips, it's not benefiting all scenarios equally. There are tons of applications that need better single threaded performance.

All things being equal, you'd be better off with a 4 core CPU than an 8 core CPU provided that they had the same overall thoroughput (i.e. a 4ghz 4 core CPU vs a 2ghz 8 core CPU of identical architecture).

It's really simple actually. If 8 cores outperform 4 cores, then more cores are preferable and that's the bottom line, end of story. Sure you can pick a couple corner cases where the i3 can outperform thuban, but i'll guarantee you that I can pick more cases that thuban demolishes i3. So because you find one or two examples where i3 is faster, you are suggesting that proves fewer cores are better? C'mon dude, i've no doubt you know better than that.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
59
91
I will agree if we only talking about one application running at any given time, but what if we run two apps simultaneously? 16 cores will be faster than 8 and what if we run 3 or four apps simultaneously etc.

AMD and Intel don’t make CPUs for Desktop/Laptop use only

Without a doubt you are correct, I am definitely addressing only those cases of multitasking single-threaded (or significantly low-threaded) applications and cases of fully loading the CPU with a single multi-thread application.

Multi-tasking multi-threaded apps takes the situation and opportunity to the next level.
 

Makaveli

Diamond Member
Feb 8, 2002
4,762
1,162
136
Either Weakboy forgot his meds or post like that on purpose. He is far too consistent with incorrect data and just bizarre post.
 

Idontcare

Elite Member
Oct 10, 1999
21,118
59
91
Oh really, this I did not know. Sorry if it came up before.. I thought the X meant HT on QX series. thanks

Do I even want to ask what you thought/think the Q stands for?

(plz no one else jump ahead and answer that, let's see what the TB comes up with...)

FWIW the X stands for eXtreme...i.e. those were the unlocked extreme edition chips.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |