Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [WCCF]AMD Bulldozer "FX-8150P Black Edition" Details Leaked, Turbo Core upto 4.7Ghz
New Posts  All Forums:Forum Nav:

[WCCF]AMD Bulldozer "FX-8150P Black Edition" Details Leaked, Turbo Core upto 4.7Ghz - Page 9  

post #81 of 100
Quote:
Originally Posted by Iceman23 View Post
I'm not holding another person's hand in understanding why it won't need 4.7 Ghz to match SB. Reread the thread again and try to comprehend. You still seem to think I'm arguing BD won't match SB

I never said that, you are just putting words in my mouth yet again. I was referring to 4.2 GHz.

I said that you can't know that the Bulldozer needs 4.2 to keep up with SB, I said for all we know, that speed could decimate SB.



And you where saying?

Quote:
Originally Posted by Iceman23

I didn't miss the memo! You are completely failing to understand. I said that right in my previous post that it was a new architecture. That's why I assumed that the new architecture would be a performance improvement, and not need 4.2GHz to match SB.
So what exactly is it that you are saying. Because as far as I'm aware your talking in riddles 90% of the time.

What I got out of the highlighted red is this: You assumed that the "Bulldozer" architecture was designed to be an improvement over their current architecture in performance, and that the new architecture shouldn't need the clock speed of 4.2 GHz to match the Sandy bridge architecture.


Edit: Stop with this comprehension crap, I am reading your posts well enough and understanding them.
My Comprehension is just fine, I have no problem understanding what my Analysis and Design or Networking, Programming, Multimedia and Web Programming teachers at Uni says and the tasks that are set.
Edited by smash_mouth01 - 6/4/11 at 6:23pm
post #82 of 100
Quote:
Originally Posted by Blameless View Post
You are implying a distinction where there is none.

BD can fit 8 cores on an 32nm AM3+ chip because of architectural improvements.

Intel cannot make 8 core 32nm SB chips and fit them in an LGA-1155 package. The cores are too damn big, the transistor count is too high.

Also, more cores is not a dead-end, it's anything but.

You are talking about per core IPC, which BD does seem like it will be lacking in, but this is becoming less relevant as multithreading becomes more and more common.
Yeah you are mostly right, I only disagree on the dead-end part.
Just upping the core count is a dead-end simply because at some point, putting more cores won't be possible for while (when we reach the SOI limit). While modifying the architecture (I can't find a better word) will always be possible.

There is also the case where a new feature appears. I'll take the GPU situation to explain my point. 10 Radeon 3870 (impossible, I know) while theoretically faster than a single 6850 can't use tesselation. Imagine something similar for CPUs. 10 cores with basic architecture VS 4 cores with advanced architecture might bear surprising results if this type of feature starts to appear.

However, everything I'm saying is very theoretical. You're right when saying that with multithreading's advent, more cores is better, but that might not be true for too long, that's what I'm implying.
post #83 of 100
Quote:
If it could significantly beat the 2600k, it would be priced much higher than 350 dollars.

If the architecture had more headroom, they would clock the thing even higher and try to put out an 800-1000 dollar chip to dethrone the 990x.
do you know anything about economics? when the amd will sell 30 milions of cpus at 350$ and get 30$ from a single cpu, this will give them 900 milions of total income. now lets take a 800$ price (not many of us can afford that expencive cpu) and sell it in less then 10000 units with a 450$ income from one cpu. it will give small 45 milions $. and fx 8150 BD will be more mainstream stuff, who knows what accualy amd will have for total enthusiast class. remember that at the beggining of bloomfields 1000$ chip was almost same bloomfield as a 280$ i7 920. 2nd thing, amd probably wants to overtake both markets by greater numbers. performance, and performance/price markets, that could be the reason for not putting noticeably higher price for bulldozers.

but in the end, we still not know true perfomance of BD. theory can by far different then real tests.

Quote:
Also, more cores is not a dead-end, it's anything but.
it is a dead end. there is nothing worse then more cores accualy. bigger chip, higher latency, bigger tdp and cooling problems, power draw, and lower performance scaling due to worse latency and all other problems with managing and sending data over the cpu. HD5k showed that nicely
Edited by prznar1 - 6/4/11 at 6:09pm
post #84 of 100
Quote:
Originally Posted by smash_mouth01 View Post
I never said that, you are just putting words in my mouth yet again. I was referring to 4.2 GHz.

I said that you can't know that the Bulldozer needs 4.2 to keep up with SB, I said for all we know, that speed could decimate SB.

And you where saying?

So what exactly is it that you are saying.
This is still going completely over your head. Read the thread again, Blameless has laid it out very clearly.
post #85 of 100
Everyone is getting bent out of shape because of the high clock speed. This is one of the better Bulldozers to be released at the end of the year correct? So that means that the 8130P@3.8 will be the one that needs to compete against SB, not the 8150P@4.2.

Also, since AMD needs to be able to undercut Intel on price, and the fact that the 2600k is closer to $300 dollars instead of $350, I think that the 8150P will need to be a bit better than the 2600k since it is not coming till the end of the year. This all sounds fine to me.

I remember reading somewhere that AMD was increasing the clock rate and increasing the IPC on their procs a while back, so 4.2 is no surprise to me. When a new architecture arrives on a die shrink, I think there are two choices:

increase CPU complexity which increases the IPC of the processor offsetting the smaller process by increasing the die size and thus heat, or increase CPU frequency, which may not lead to a die size increase, but will lead to an increase in TDP and heat anyway.

The latter is the route AMD chose, and is fine since increasing complexity is difficult to achieve and has to end somewhere. So I dont see why there is so much crying over the fact.
Edited by Quantium40 - 6/4/11 at 6:14pm
Black Hole
(16 items)
 
  
Black Hole
(16 items)
 
  
post #86 of 100
CORE COUNT AND CLOCK SPEED ARE SO IMPORTANT!

I never need to see benchmarks.
    
CPUMotherboardGraphicsRAM
i7-2600k 4.5GHz @ 1.32V Asus P8P67 Pro EVGA GTX 580 Mushkin 2133 9-10-9-24 
Hard DriveHard DriveCoolingOS
Samsung 840  WD Black Silver Arrow Windows 7 
MonitorKeyboardPowerCase
Dell U2211H Rosewill RK-9000BR Seasonic X750 HAF X 
MouseAudio
Razer Lachesis Grado HF2 
  hide details  
    
CPUMotherboardGraphicsRAM
i7-2600k 4.5GHz @ 1.32V Asus P8P67 Pro EVGA GTX 580 Mushkin 2133 9-10-9-24 
Hard DriveHard DriveCoolingOS
Samsung 840  WD Black Silver Arrow Windows 7 
MonitorKeyboardPowerCase
Dell U2211H Rosewill RK-9000BR Seasonic X750 HAF X 
MouseAudio
Razer Lachesis Grado HF2 
  hide details  
post #87 of 100
Quote:
Originally Posted by Blameless View Post


You, and people in your situation, are not the only potential AMD customers out there.

I care how BD will stack up, because I'm going to want some new CPUs soon, and I already have some that are close enough in performance to SB for SB to not really be much of an upgrade for me.
Oh don't get me wrong, I know that there are a full spectrum of buyers out there, but if only people would wait it would be well worth the wait, IMO. I don't like the posts where people cry about BD not being on retail shelves yesterday and then threatening to goto Intel. I know it is important to some how it stacks up against Intel, but there are others that don't care about that angle of the equation.
BULLDOZER!!!
(14 items)
 
  
CPUMotherboardGraphicsRAM
8120FX Gigabyte GA-990FXA-UD7 2ea MSI r9-270x 16gb G-Skill Snipers 
Hard DriveOptical DriveCoolingOS
WD 3x1tb greens, 1x1tb black Adata S599 120gb LG DVDR/RW, LG Blueray CM V8 Win7 64bit 
MonitorKeyboardPowerCase
LG2762D Microsoft Antec TPN-750 CM HAF 932 
MouseMouse Pad
Microsoft X5 Generic 
  hide details  
BULLDOZER!!!
(14 items)
 
  
CPUMotherboardGraphicsRAM
8120FX Gigabyte GA-990FXA-UD7 2ea MSI r9-270x 16gb G-Skill Snipers 
Hard DriveOptical DriveCoolingOS
WD 3x1tb greens, 1x1tb black Adata S599 120gb LG DVDR/RW, LG Blueray CM V8 Win7 64bit 
MonitorKeyboardPowerCase
LG2762D Microsoft Antec TPN-750 CM HAF 932 
MouseMouse Pad
Microsoft X5 Generic 
  hide details  
post #88 of 100
Quote:
Originally Posted by Blameless View Post
Intel back in the NetBurst days, was racing to 4GHz, because AMD's ~2.5GHz Athlon 64s were kicking the crap out of their 3.6-3.73GHz CPUs.

There has been something of a reversal of this situation in the last few years, except that Intel has the resources to keep nearly a full process shrink ahead and maintain a clock and IPC lead.

Many people, including myself, have been hoping that BD would change this.
This.

AMD needs to have a wicked new architecture. \\discussion
Nightwish
(14 items)
 
  
CPUMotherboardGraphicsRAM
i7 950 @ 4.083GHz 1.376v BIOS 1.352v CPU-Z EVGA 141-BL-E757-RX EVGA GTX 460 1GB G.Skill 6GB DDR3 1600 
Hard DriveOptical DriveCoolingOS
500GB WD Green, 1TB Samsung F3, 2x 80gb Seagate... LG 10x BluRay Burner, LG DVDRW Thermalright LAPPED true Windows 7 Ultimate 64bit 
MonitorKeyboardPowerCase
Samsung 20" LCD Saitek Eclipse Thermaltake 750w Cosmos S 
Mouse
CoolerMaster Sentinel 
  hide details  
Nightwish
(14 items)
 
  
CPUMotherboardGraphicsRAM
i7 950 @ 4.083GHz 1.376v BIOS 1.352v CPU-Z EVGA 141-BL-E757-RX EVGA GTX 460 1GB G.Skill 6GB DDR3 1600 
Hard DriveOptical DriveCoolingOS
500GB WD Green, 1TB Samsung F3, 2x 80gb Seagate... LG 10x BluRay Burner, LG DVDRW Thermalright LAPPED true Windows 7 Ultimate 64bit 
MonitorKeyboardPowerCase
Samsung 20" LCD Saitek Eclipse Thermaltake 750w Cosmos S 
Mouse
CoolerMaster Sentinel 
  hide details  
post #89 of 100
So guys... hate to ruin your day but
ORIGINAL ARTICLE from China, TRANSLATED
Quote:
Here is the FX-8150P know the source of the specifications. Originally published as the April 1, so the authenticity to be confirmed.
post #90 of 100
Quote:
Originally Posted by Iceman23 View Post
This is still going completely over your head. Read the thread again, Blameless has laid it out very clearly.
But this has nothing to do with Blameless or what he said, this is directed at what you said nothing more or less..

I was not arguing a point you where, and to be doing it off of something that has not been verified. So stop changing the subject.


New news to light, it looks to be an April fools.
Edited by smash_mouth01 - 6/4/11 at 6:30pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
This thread is locked  
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [WCCF]AMD Bulldozer "FX-8150P Black Edition" Details Leaked, Turbo Core upto 4.7Ghz