Rumour: Bulldozer 50% Faster than Core i7 and Phenom II.

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.

veri745

Golden Member
Oct 11, 2007
1,163
4
81
Yea I bet grandma is really going after an 8 core processor.

What about the 95% of people who never use a program that utilizes more than 4 cores?

Clever marketing by AMD, but not a real-world situation.

Not only did people say the same thing about quads when Ph I came out, they said the same thing about dual cores when the A64 X2 came out.
 

BLaber

Member
Jun 23, 2008
184
0
0
A BD module has 2 cores but , comparing 4 BD module to 4 I7 core + 4 Ht threads is not apples to apples comparison because of huge difference in design philosophy of both the BD & I7 processors.

I guess the best way to compare performance is to compare die size of 1 BD module vs die size of 1 I7 core + ht thread for that core.
 

Wreckage

Banned
Jul 1, 2005
5,529
0
0
If Grandma uses IE8+ or Chrome then she is using a program that does utilize more than 4 cores. Especially if you mix in running apps/video with Flash, Silverlight, or Java inside the browser.


Maybe she should switch to Firefox which runs just fine on an ole P4.


Seriously if you are trying to sell someone on an 8 core CPU for the internet.... I'm guessing you are just being sarcastic.
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
No, this is not true, they are all full cores. Every integer core has its own FP unit (actually a more powerful FMAC.)

It will be 8 cores, 8 threads.

Thanks for clearing that up. So an 8 core CPU is 50% faster than a 4 core CPU with HT. Clearly that is good news for AMD as it will be a nice performance boost. And even though I am happy with Intel, I would like to see more price wars.

But, based on those numbers (if true), means that an 8 core BD would be on a performance par with a 6 core SB (without HT enabled).

Now the question is will that 8 core BD be much cheaper than a 6 core SB.
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
A BD module has 2 cores but , comparing 4 BD module to 4 I7 core + 4 Ht threads is not apples to apples comparison because of huge difference in design philosophy of both the BD & I7 processors.

I guess the best way to compare performance is to compare die size of 1 BD module vs die size of 1 I7 core + ht thread for that core.

Or compare 1 BD core to 1 Intel core with HT turned off. At least if you want to see the IPC differences.
 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
Ya AMD said the same about Phenom I didn't they . Slides are still in those old topics.

Hey, could you ask your hero 'Bob for apples' why hardly any vendors want anything to do with oak trail? I thought you and big bad Bob claimed it was the second coming of Christ.
 

HW2050Plus

Member
Jan 12, 2011
168
0
0
You seem very certain about something that hasn't been released, much less benchmarked.
Right. But the architecture has been released by AMD and there are some clear design informations, showing that Bulldozer will double the core count (simplified), in addition it is known that Bulldozer will run at 4-5 GHz (4.5 GHz would be a good guess which comes from 3.2 GHz * 23 / 17 [pipeline design FO4 count] + 5% [from 32 nm]).

It is quite a simple assumption that double of cores will give 70-100% gain. That is something we know already. So nothing magic with that. That is why one can be certain at least if overall performance is regarded. If it is better in a certain application or another I really can't tell. That and the exact performance gain within that range has to be shown by benchmarking the released product. As current rumors indicate, it would be at the high end of this range, but that is just based on rumors.
 
Last edited:

OBLAMA2009

Diamond Member
Apr 17, 2008
6,574
3
0
er, uh, you dont fire your ceo for producing a chip that outperforms the competition by 50%
 

HW2050Plus

Member
Jan 12, 2011
168
0
0
I guess the best way to compare performance is to compare die size of 1 BD module vs die size of 1 I7 core + ht thread for that core.
The beauty is that they are about the same.

Just look at IBM Power 7 processor. AMD is not first using this "module technique" and not the last. Intel will be heavily forced to make this design step as well (as they followed with 64 Bit/integrated memory controller/processor interconnect/L3 cache etc.).
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
Right. But the architecture has been released by AMD and there are some clear design informations, showing that Bulldozer will double the core count (simplified), in addition it is known that Bulldozer will run at 4-5 GHz (4.5 GHz would be a good guess which comes from 3.2 GHz * 23 / 17 [pipeline design FO4 count] + 5% [from 32 nm]).

It is quite a simple assumption that double of cores will give 70-100% gain. That is something we know already. So nothing magic with that. That is why one can be certain at least if overall performance is regarded. If it is better in a certain application or another I really can't tell. That and the exact performance gain within that range has to be shown by benchmarking the released product. As current rumors indicate, it would be at the high end of this range, but that is just based on rumors.

Pure speculation. Please provide evidence and/or links to support these facts...HIGHLY doubtful of a 70-100% gain. Not going to happen.
 

Makaveli

Diamond Member
Feb 8, 2002
4,762
1,161
136
Its all heresay right now no one has any concrete numbers and most are working with theories!
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
1090T @ 4.0ghz easily got 7.0 in Cinebench R11.5.
980X @ 3.33ghz = 8.95
2600k @ 3.4ghz = 6.88 (Bit-tech)
i7-950 @ 3.06ghz = 5.57 (Bit-tech)

If we assume perfect "core" scaling from 6 core Stars Phenom II to 4 core/8 modules BD, that's 33.3% increase. I am going to be conservative and say it will only be a 20% increase. Let's assume BD will net at least a 10% increase in IPC over Stars (again on the conservative end).

On 45nm, 1100T is already at 3.3ghz with 3.7ghz Turbo. This means getting a 3.8-4.0ghz BD at launch is not out of the question on 32nm.

3.8ghz BD x 20% increase in cores x 10% increase in IPC = 5.0ghz Phenom II X6.

This would get us closer to 8.8 on Cinebench R11.5 with a stock BD.

Here is your 50% increase over i7-950 without breaking a sweat in a multi-threaded app and performance right near 980X.

And these are very conservative estimates on my part.
 

nyker96

Diamond Member
Apr 19, 2005
5,630
2
81
I think if all the speculation on the street is true, then each BD core is roughly equal to 3/4 of Nahalem core. roughly, an 8core BD = 6core 980X roughly. maybe slightly less. so I conjecture that since SB is about 15% stronger IPC than Nehalem core, I'd say
1 BD core = 0.6 to 0.65 SB core.

so ::
an 8 core BD = 5 core SB
6 core BD = 4 core SB
4 core BD = 2.5 core SB
 

SolidSnake42

Senior member
Feb 9, 2010
263
0
0
I'd just love to see what this bad boy can do when all the sites start doing their own tests. I've used both AMD and Intel in the past, so I have no biasness. I would love to see AMD get up there though and be competitive, neck and neck with Intel to make things fun, interesting and provide more and better choices. I'm used to being one sided for console wars, but when it comes to processors, I just go with what better suits my performance needs and even if I pick one over the other, I want the other to do well too. But if we can expect these kind of given performance stats, then give AMD props to really try to push themselves to do what they can to go toe-to-toe with Intel.
 

Daedalus685

Golden Member
Nov 12, 2009
1,386
1
0
er, uh, you dont fire your ceo for producing a chip that outperforms the competition by 50%

His removal/firing/resignation/whatever likely had nothing to do with any single product at all and more to do with business strategy differences between him and the board.
 

Matrices

Golden Member
Aug 9, 2003
1,377
0
0
Whoever can open up the bottleneck of GTX 580 Tri-SLI at 2500x1600 (or equivalent-level performance) will win my money. At that resolution, and with those cards, the best i7 processor is still a hard bottleneck in a number of games, according to GURU3D.com.
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
His removal/firing/resignation/whatever likely had nothing to do with any single product at all and more to do with business strategy differences between him and the board.

Exactly. He thought BD should have been released in 2009. The board thought 2011 was better.
 

Edrick

Golden Member
Feb 18, 2010
1,939
230
106
Whoever can open up the bottleneck of GTX 580 Tri-SLI at 2500x1600 (or equivalent-level performance) will win my money. At that resolution, and with those cards, the best i7 processor is still a hard bottleneck in a number of games, according to GURU3D.com.

Including a 980X and 2600k?
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Whoever can open up the bottleneck of GTX 580 Tri-SLI at 2500x1600 (or equivalent-level performance) will win my money. At that resolution, and with those cards, the best i7 processor is still a hard bottleneck in a number of games, according to GURU3D.com.

Send one of those 580s to me. And your bottleneck will be solved. :wub:
 

Makaveli

Diamond Member
Feb 8, 2002
4,762
1,161
136
considering 4 current intel cores matches 6 amd cores who really made the poor design choice?

SB-E 6 - 8 core version will be coming out around the same time and this will be competition for BD not socket 1155.

christmas 2011 will be an great time to build a new system.
 
Status
Not open for further replies.
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/    |